Original subtitles

the rise of AI has been largely driven

by one too in AI called machine learning

in this video you learn what is machine

learning so that by the end you hope

you'll be able to start thinking how

machine learning might be applied to

your company or to your industry the

most commonly used type of machine

learning as a type of AI that learns a

to be or input to output mappings and

this is called supervised learning let's

see some examples if the input a is an

email and the output B you want is this

email spam one on 0 1 then this is the

core piece of AI used to build a spam

filter or if the input is an audio clip

and the a eyes job is output D text

transcript dentists is speech

recognition more examples if you want to

input English and have it outputs a

different language Chinese Spanish

something else then this is machine

translation or the most lucrative form

of supervised learning of this type of

machine learning maybe online

advertising where all the large online

ad platforms have a piece of AI that

inputs some information above an ad and

some information about you and tries to

figure out will you click on this ad or

not and by showing you the answer you

most likely click on this turns out to

be very lucrative maybe not the most

inspiring application but certainly

having a huge economic impact today or

if you want to build a self-driving car

one of the key pieces of AI is in the

IDE that teaches input an image and some

information from the radar or from other

sensors and outputs the position of

other costs so your self-driving car can

avoid the other cause or in

manufacturing I've actually done a lot

of work in manufacturing where you take

as input a picture of something you've

just manufacture such as a picture of a

cell phone coming off an assembly line

this is a picture of a phone another

picture taken by a phone and you want to

output is there a scratch was there a

dancer as some other defects on this

thing you've just manufactured and this

is visual inspection which is helping

manufacturers

reduce or prevent defects in the things

that they're making this type of AI

called supervised learning just learns

input to output or a to be mappings and

on one hand input output ABB seems quite

limiting but when you find the right

application scenario this can be

incredibly valuable now the idea of

supervised learning has been around for

many decades but that's really taken off

in the last few years why is this where

my friends asked me hey Andrew why is

supervised learning is taking off now

there's a picture I draw for them and I

want to show you this picture now and

you may be able to draw this picture for

others that ask you the same question as

well let's say on the horizontal axis

you plot the amount of data you have

Berta's so for speech recognition this

might be the amount of audio data and

transcripts you have in a lot of

industries the amount of data you have

access to has really grown over the last

couple decades thanks to the rise in the

Internet the rise of computers a lot of

what used to be say pieces of paper are

now instead recorded on a digital

computer so we've just been getting more

and more and more data now let's say on

the vertical axis you plot the

performance of an AI system it turns out

that if you use a traditional AI system

then the performance would grow like

this then as you feed it more data as

performance gets a bit better but beyond

a certain point it did not get that much

better so that if your speech

recognition system did not get that much

more accurate or your online advertising

system didn't get that much more

accurate than showing the most relevant

ads even as you showed it more data AI

has really taken off recently due to the

rise of neuro networks and deep learning

how to find these terms more precisely

in later videos so don't worry too much

about what it means but now but with

modern AI with neural networks and deep

learning what we saw was that if you

train a small neural network then to

perform

you may have heard that data is really

important for building AI systems but

what is data really let's take a look

let's look at an example of a table of

data which we also call a data set if

you're trying to figure out how to price

houses that you trying to buy or sell

you might collect a data set like this

and this can be just a spreadsheet like

an Excel spreadsheet of data where one

column is the size of the house say in

square feet or square meters and the

second column is the price of the house

and so if you're trying to build an AI

system a machine learning system to help

you set prices for houses or figure out

of a houses price appropriately you

might decide that the size of the house

is a and the price of the houses B and

have an AI system learn this input to

output or a to be mapping now rather

than just pricing a house based on the

size you might say well let's also

collect data on the number of bedrooms

of this house in that case a can be both

of these first two columns and B can be

just the price of the house so given a

table of data given the data set there's

actually up to you up to your business

use case to decide what is a and what is

B data is often unique to your business

and this is an example of a data set

that a real estate agency might have if

they trying to help price

pulses and it's up to you to decide what

is a and what is B and how to choose

these definitions of a and B to make it

valuable for your business as another

example if you have a certain budget and

you want to decide what is the size of

house you can afford then you might

decide that the input a is how much does

someone spend and B is just the size of

the home

you might have heard terminology from AI

such as machine learning or data science

or neural networks or deep learning what

do these terms mean in this video you

see what is this terminology of the most

important concepts of AI so that you

will speak with others about it and

start thinking how these things could

apply in your business let's get started

let's say you have a housing data set

like this with the size of house number

bedrooms and Rabab rooms what are the

houses newly renovated as well as the

price if you want to build a mobile app

to help people price houses so this

would be the input a and this would be

the outputs B then this would be a

machine learning system in particular

it'd be one of those machine learning

systems that learns inputs to outputs or

a to be mappings so machine learning

often results in a running AI system so

there's a piece of software that any

time of day any time of night you can

automatically input a these properties

of a house and a plus B so if you have

an AI system running serving dozens or

hundreds of thousands of millions of

users that's usually a machine learning

system in contrast here's something else

you might want to do which is to have a

team analyze your data set in order to

gain insights so a team might come up

with a conclusion like hey did you know

if you have two houses of a similar size

of a similar square footage if the house

has three bedrooms then they cost a lot

more than the house of two bedrooms even

if the square footage is the same or did

you know that newly renovated homes have

a fifteen percent premium and this could

help you make decisions such as given a

similar square footage do you want to

build a two bedroom or a three bedroom

size in order to maximize value or is it

worth in investments to renovate a home

in the hope that the renovation

increases the price you can sell a house

for so these would be examples of data

science projects where the output of a

data science project is a set of

insights that can help you make business

decisions such as what type of house to

build or whether to invest

in renovation the boundaries between

these two terms machine learning and

data science are a little bit fuzzy and

these terms are not used consistently

even in industry today but what I'm

giving here is maybe the most commonly

used definitions of these terms but you

will not find universal adherence to

these definitions so formalize these two

notions a bit more machine learning is

the field of study that gives computers

the ability to learn without being

explicitly programmed this is a

definition by author Samuel many decades

ago after Samuel was one of the pioneers

of machine learning who was famous for

building a checklist playing program

that could play checkers even better

than he himself the inventor could play

the game so a machine learning project

will often result in a piece of software

that runs that outputs be given a in

contrast data science is the signs of

extracting knowledge and insights from

data and so the output of a data science

project is often a slide deck Department

presentation that summarizes conclusions

for executives to take business actions

or that summarizes conclusions for a

product team to decide how to improve a

website let me give an example of

machine learning versus data science in

the online advertising industry today

the large high platforms all have a

piece of AI that quickly tells them

what's the ad you are most likely to

click on so that's a machine learning

system and this turns out to be

incredibly lucrative AI system the

inputs information about you and about

the ad and outputs will you click on

this or not these systems are running

24/7 and these are machine learning

systems that drive ad revenue for these

companies so there's a piece of software

that runs in contrast I've also done

data science projects in the online

advertising industry if analyzing data

tells you for example that the travel

industry is not buying a lot of ads but

if you send more sales people to sell

ads the travel companies you could

convince them to use more advertising

then that would be an example of a data

science project and the data science

conclusion

the results in the executives deciding

to ask the sales team to spend more time

reaching out to the travel industry so

even in one company you may have

different machine learning and data

science project spoke for which can be

incredibly valuable you have also heard

of deep learning so what does deep

learning let's say want to predict

housing prices you want to price houses

so you have an input that tells you the

size of house number of bedrooms and

bathrooms and where this newly renovated

one of the most effective ways to priced

houses given this input a will be

defeated this thing here in order to

have it output the price this big thing

in the middle is called a neural network

and sometimes we also call it an

artificial neural network and that's the

distinguish it from the neural network

that is in your brain so the human brain

is made up of neurons and so when we say

artificial neural network that's just

emphasize that this is not the

biological brain but it says a piece of

software and what a neural network does

we're not official neural network does

is take this input a which is all of

these whole things and then output B

which is the estimated price of the

house now in a later optional video this

week I'll show you more what this

artificial neural network really is but

all of human cognition is made up of

neurons in your brain

passing electrical impulses positive low

messages each other and when we draw a

picture of an artificial neural network

there's a very loose analogy to the

brain and these little circles are

called

artificial neurons or just neurons for

short that also passes in neurons to

each other and this big artificial

neural network is just a big

mathematical equation that tells it

given the inputs a how do you compute

the price B in case it seems like there

are a lot of details here don't worry

about it we'll talk more about these

details later but the key takeaways are

that a neural network is a very

effective technique for learning a to be

your input output mapping

and today determines neural network and

deep learning are used almost

interchangeably they mean essentially

the same thing many decades ago this

type of software was called a neural

network but in recent years we found

that you know deep learning was just a

much better sounding brand and so that

thought better versus the term that's

been taking off recently so what do new

networks or artificial neural networks

have to do with the brain it turns out

almost nothing new networks were

originally inspired by the brain but the

details of how they work are almost

completely unrelated to how biological

brains work so I choose very courses

today about making any analogies between

artificial neural networks and the

biological brain even though there was

some loose inspiration there so AI has

many different tools in this video you

learned about what a machine learning

and data science and also what is deep

learning and was it neural network you

might also hear in the media other

buzzwords like unsupervised learning

wrinkles learning graphic novels

planning knowledge drop and so on and

you don't need to know what all of these

other terms mean but these are just

other tools for getting AI systems to

make computers act intelligent you know

try to give you a sense of what some of

these terms mean in later videos as well

but the most important tools that I hope

you know about are machine learning and

data science as well as deep learning

the neural networks which are a very

powerful way to do machine learning and

sometimes data science if we were to

draw a Venn diagram showing how all

these concepts fit together and this is

what it might look like AI is this huge

set of tools for making computers behave

intelligently of AI the biggest subset

is very tools from machine learning but

AI does have other tools than machine

learning such as some of these buzz

words are listed at the bottom and of

machine learning the part of machine

learning that's most important these

days is

neural networks or deep learning which

is very powerful set of tools for

carrying out supervised learning or a to

be mappings as well as some other things

but they're also other machine learnings

who

that are not just deep learning tools so

how does data science fit into this

picture there is inconsistency in how

two terminologies use some people will

tell you the designs is a subset of AI

some people will tell you AI is a subset

that they design so it depends a bit on

who you ask

but I would say that data science is

maybe a cross-cutting subset of all of

these tools that uses many tools from AI

machine learning and deep learning but

has some other separate tools as well

that solves a very set of important

problems in driving business insights in

this video you saw what is machine

learning where the state designs and

what is deep learning and neural

networks I hope this gives you a sense

of the most common and important

terminology using AI and you can start

thinking about how these things might

apply to your company now what does it

mean for a company to be good at AI

let's talk about that in the next video

you

what makes a company good at AI and

perhaps even more importantly what would

it take for your country to become great

and using AI I had previously led the

Google brain team and by deuce AI group

which I respectively helped Google and

Baidu become great AI companies so what

can you do for your company

thus a lesson had learned washing the

rise of the internet that I think would

be relevant to how all of us navigate

the rise of AI let's take a look a

lesson we learned from the rise of the

internet was that if you take your

favorite shopping mall so you know my

wife and I sometimes shop at Stanford

Shopping Center and you put a website

for the shopping mall maybe sell things

on the website that by itself does not

turn the shopping mall into an Internet

company in fact a few years ago I was

speaking with the CEO of a large retail

company who said to me hey Andrew I have

a website I sell things on the website

amazon has a website Amazon sells things

a website is the same thing but of

course it wasn't in the shopping mall

with a website isn't the same thing as a

first-class Internet company so what is

it that defines an Internet company if

it isn't just whether or not you sell

things on the website I think an

Internet company is a company that does

the things the internet lets you do

really well for example we engage in

pervasive AP testing meaning we

routinely throw up two different

versions of web site and see which one

works better because we can and so we

learn much faster whereas in a

traditional shopping mall you know very

difficult to have two shopping malls in

two parallel universes and you can only

maybe chase things around every quarter

every six months Internet companies tend

to have very short iteration time so you

can ship a new product every week or

maybe even every day because you can

wear as a shopping mall can be

redesigned and we are protected only

every several months Internet companies

also tend to push decision making down

from the CEO to the engineers and to

other specialized roles such as the

product managers

this is in contrast to a traditional

shopping mall where you can maybe have

the CEO just decide all the key

decisions and then just everyone does

what the CEO says and it turns out that

traditional model doesn't work in the

internet error because only the

engineers and other specialized roles

like product managers know enough about

the technology and the product and the

users to make great decisions so these

are some of the things that internet

companies do in order to make sure they

do the things that the internet doesn't

do really well this is a lesson we learn

from the internet error how about the AI

error

I think that today you can take any

company and haven't used a few neural

networks or few deep learning algorithms

that by itself does not turn the company

into an AI company instead what makes a

great AI company is sometimes an AI

first company is are you doing the

things that AI lets you do really well

for example AI companies are very good

at strategic data acquisition this is

why many of the large consumer tech

companies may have three products that

do not monetize and it allows them to

acquire data that they can monetize

elsewhere so let strategy teams where we

would deliberately launch products that

do not make any money just for the sake

of data acquisition and thinking through

how to get data is a key part of the

great AI companies a company sends up

unified data warehouses if you have 50

different databases or 50 different data

warehouses under the control of 50

different vice-presidents then they'll

be impossible for an engineer to get the

data into one place so that they can

connect the dots and swap the patterns

so many great our companies have

preemptively invested in bringing the

data together into a single data

warehouse to increase the odds that the

teams can connect the dots

subjective cause to privacy guarantees

and also to data regulations such as GDP

are in Europe our companies are very

good at spotting automation

opportunities we're very good at seeing

oh let's insert the supervised learning

Albert

and have a a to be mapping here so that

we don't have to have people do these

tasks instead we can automate it yeah I

companies also have many new roles such

as the MLE or machine learning engineer

and new ways of dividing up toss among

different members of the team so for

company to become good at AI means

architecting the company to do the

things that AI makes it possible to do

really well now for a company to become

good at AI does require a process in

fact 10 years ago Google and Baidu as

well as companies like Facebook and

Microsoft that was now the part of we're

not great AI company is the way that

they are today so how can a company

become good at AI it turns out that

becoming good at AI is not a mysterious

magical process instead there is a

systematic process through which many

companies almost any big company can

become good at AI this is the five-step

AI transformation playbook that I

recommend to companies that want to

become effective and using AI I'll give

a brief overview of the playbook here

and then go into detail in a later week

step one is to execute pilot projects to

gain momentum so just do a few small

projects to get Ben a sense of what AI

can and cannot do and get a better sense

of what doing an AI project feels like

and this you could do in-house or you

can also do with an outsource team but

eventually you then need to do step 2

which is to build an in-house AI team

and provide broad AI training not just

to the engineers but also to the

managers division leaders and executives

and how to think about AI after doing

this so as you're doing this you have a

better sense of what AI is and then it's

important for many companies to develop

an AI strategy and finally to align

internal and external communications so

that all your stakeholders from

employees customers and investors are

aligned with how your company is

navigating the rise of AI the AI has

created tremendous value in the software

industry and will continue to do so it

will also create tremendous value

outside the software industry

if you can help your company become good

at AI I hope you can play a leading role

in creating a lot of this value in this

video you saw what is it that makes a

company a good AI company and also

briefly the AI transformation playbook

which they're going to much create a

detail on in a later week as a roadmap

for helping companies become great at AI

if you're interested there is also

published online an AI transformation

playbook that goes into these five steps

in greater detail but you see more of

these in the later league as well now

one of the challenges of doing a are

projects such as the pilot project in

step one is

in this video and the next video I hope

to help you develop intuition about what

a I can and cannot do in practice before

I commit to a specific AI project I'll

usually have either myself or engineers

do technical diligence on the project to

make sure that it is feasible this means

look in the data look at the input and

output a and B and just thinking through

if this is something a I can really do

what I've seen unfortunately is that

some CEOs can have an overinflated

expectation of AI and can ask engineers

to do things that today's AI just cannot

do one of the challenges is that the

media as well as the academic literature

tends to only report on positive results

of success stories using AI and we see a

string of success stories and no failure

stories people sometimes think AI can do

everything and unfortunately that's just

not true so what I want to do in this

and the next video is to show you a few

examples of what today's AI technology

can do but also what it cannot do and I

hope that this will help you hone your

intuition about what might be more or

less promising projects to select for

your company previously you saw this

list of AI applications from span 14 to

speech recognition to machine

translation and so on one imperfect rule

of thumb you can use to decide what

supervised learning may or may not be

able to do is that pretty much anything

you can do with a second of thought we

can probably now or soon automates using

supervised learning using this

input/output mapping so for example in

order to determine the position of other

costs you know that's something that you

can do with less than a second in order

to tell if a phone is strache you can

look at it and you can kind of tell in

less than a second in order to

understand at least transcribe what was

said you know doesn't take that many

seconds of thought and while this is an

imperfect rule of thumb it maybe gives

you a way to quickly think of some

examples of tasks that AI systems can do

whereas in contrast something that AI

today cannot do would be to analyze a

market and write

50 page report the human cannot write a

50 page market analysis report in a

second and it's very difficult and he's

I don't know and I don't think any team

in the world today knows how to get an

AI system to do market research and run

an extended market report either I found

that one of the best ways the whole

intuition is a look at concrete examples

so let's take a look at a specific

example relating to customer support

automation let's see you run a website

there sell things so an e-commerce

company and you have a Customer Support

Division that gets an email like this

the tour arrived two days later I wasn't

going to give it to my niece for her

birthday can I return it if what you

want is an AI system that looks at this

and decides this is a refund request

so let me route it to my refund

department then I was saying you have a

good chance of building an AI system to

do that the AI system would take as

input the customer checks what the

customer emails you and it would output

this is a refund request or is this a

shipping problem or is it a other

requests in order to route this email to

the most appropriate part of your

customer support center so the input aid

is at X and the output B is one of these

three outcomes there's a refund or a

shipping problem or shipping query or is

it a different request so this is

something that AI today can do here's

something a act today cannot do which is

if you want AI to input an email and

automatically generate a response like

oh sorry here that I hope you needed a

good birthday yes we can help work and

so on

so for an AI to output a complicated

piece of text like this today is very

difficult by today's standards of AI and

in fact to even empathize about the

birthday of your knees that is very

difficult to do for every single

possible type of email you might receive

now what would happen if you were to use

a machine learning tool like a deep

learning algorithm to try to do this

anyway so let's say you try to get an AI

system to input the user's email and

open a to the three-paragraph

empathetic and appropriate response and

let's say that you have a mother size

data set you're like

thousand examples of user emails and

appropriate responses it turns out if

you run an AI system on this type of

data on a small data set like a thousand

examples this may be the performance you

get which is if a user emails my box is

damaged you'll say thank you for email

and it says whether rather review thank

you email what's written policy thank

you very much but the problem we're

building this type of AI is that with

just a thousand examples there's just

not enough data for an AI system to

learn how to write to the three

paragraph appropriate and empathetic

responses so it may end up just

generating the same very simple response

like thank you very mad what the

customer is sending you another thing

that could go wrong another way from the

assistance of fail is if it generates

gibberish such as whether my boss

arriving and it says thank yes now

you're kind of gibberish and this is a

hard enough problem that even with ten

thousand or a hundred thousand email

examples I don't know if that would be

enough data for an AI system to do this

well the rules for what AI can it cannot

do are not hard and fast and I usually

end up having to ask engineering teams

to sometimes spend a few weeks doing

deep technical diligence to decide for

myself if a project is feasible but to

hone your intuitions to help you quickly

filter feasible and not feasible

projects here are a couple of other

rules of thumb about what makes a

machine learning problem easier or more

likely to be feasible one learning a

simple concept is more likely to be

feasible well what is a simple concept

mean there's no formal definition of

that but this is something that takes

you less than a second of mental thought

or a very very small number of seconds

of mental thought to come up with a

conclusion then that would lean to

whether it being a simple concept so

you're looking outside the window of a

self-driving car to spot the other calls

that would be a relatively simple

concept

whereas how to write an empathetic

response to a complicated user

complaints that would be less of a

simple concept

seconds a machine learning problem is

more likely to be feasible if you have

lots of data available and here our data

means both the input a and the output B

that you want the AI system to have in

your a to be input-output mapping so for

example in the customer support

application the input a would be

examples of emails from customers and B

could be labeling each of these custom

emails as to whether is a refund request

or a shipping query or some other

problem one of the outcomes that if you

have thousands of emails with both a and

B then the odds of you being with build

a machine learning system to do that

would be pretty good AI is the new

electricity and is transforming every

industry but there's also not magic and

you can't do everything under the Sun I

hope that this video started to help you

hone your intuitions about what it can

and cannot do and increase the odds of

your selecting feasible and valuable

projects for maybe your teams to try

working on in order to help you continue

developing your intuition I would like

to show you more examples of what AI can

and cannot do

let's go into the next video

one of the challenges of becoming good

at recognizing what AI can and cannot do

is that it does take seeing a few

examples of concrete successes and

failures of AI and if you work on an

average of say one nu AI project a year

then to see three examples would take

you the years of work experience and

that's just a long time whether how to

do both in the previous video and in

this video is to quickly show you a few

examples of AI successes invidious or

what it can and cannot do so that in a

much shorter time you can see multiple

concrete examples to help hone your

intuition and select valuable projects

so let's take a look at a few more

examples let's say you're building a

self-driving car here's something that

AI can do pretty well which is to take a

picture of what's in front of your car

and maybe just using camera may be using

other senses as well such as radar or

lidar and then to figure out what is the

position or where are the other costs so

this would be an AI where the input a is

a picture of what's in front of your car

or maybe both a picture as well as radar

and other sensor readings and the output

B is where are the other costs and today

the self-driving car industry has

figured out how to collect enough data

and has pretty good algorithms so doing

this reasonably well so that's what a AI

today can do here's an example of

something that today's AI cannot do and

this would be very difficult using

today's AI which is to input a picture

and outputs the intention or whatever

the human is trying to gesture at your

car so here's a construction worker

holding out a hand to ask your car to

stop here's a hitchhiker trying to wave

a car over here's a bicyclist raising

the left hand to indicate that they want

to turn left and so if you were to try

to build a system to learn an ADA be

mapping where the input a is a short

video of a human gesturing at your car

and the output B is what's the intention

what does this person want that today is

very difficult to do part of the problem

is that the number of ways people

gesture at you is very very large

imagine all the hand gestures someone

could conceivably use awesome

slow down or go I'll stop the number of

ways that people could gesture at you is

just very very large and so it's

difficult to collect enough data from

enough thousands or tens of thousands or

different people gesturing at you and

all of these different ways to capture

at the richness of human gestures so

learning from a video to what this

person one since I share somewhat

complicated concept and that even people

have a hard time figuring out sometimes

what someone waving at your car wants

and then second because this is a safety

critical application you would want an

AI that is extremely accurate in terms

of figuring out there's a construction

worker want you to stop or does he or

she want you to go and that makes it

harder for an AI system as well and so

today if you collect just say ten

thousand pictures of other cars many

teams would be able to build an AI

system that at least has a basic

capability at detecting other cars

in contrast even if you collect pictures

or videos of ten thousand people is

quite hard to track down ten thousand

people waving at your car even with that

data set I think it's quite hard today

to build an AI system to recognize human

intention from their gestures and at the

very high level of accuracy needed in

order to drive safely around these

people so that's why today many sub

driving car teams have some components

for detecting other cars and they do

rely on their technology to drive safely

but very few self-driving car teams are

trying to count on an AI system to

recognize a huge diversity of human

gestures and Counting just on that they

drive safely around people let's look at

one more example say you want to build

an AI system to look at x-ray images and

diagnose pneumonia so all of these are

chest x-rays so the input a could be the

x-ray image and the output B can be the

diagnosis does this patient have

pneumonia or not so that's something

that a I can do something that a I

cannot do would be to diagnose pneumonia

from ten images of a medical textbook

chapter explaining pneumonia

a human can look at a small set of

images maybe just a few dozen images and

read a few paragraphs from a medical

textbook and start to get a sense but I

actually don't know given a medical

textbook what is a and what is B or how

to really pose this as an AI problem

that I know how to write a piece of

software to solve if all you have is

just ten images and a few paragraphs of

text that explain what pneumonia and a

chest x-ray looks like whereas a young

medical doctor might learn quite well

reading a medical textbook and just

looking at you know maybe dozens of

images in contrast an AI system isn't

really able to do that today

so summarize here are some of the

strengths and weaknesses of machine

learning machine learning tends to work

well when you're trying to learn a

simple concept such as something that

you could do with less than a second of

mental thought and when there's lots of

data available machine learning tends

were poorly when you're trying to learn

a complex concept from small amounts of

data a second underappreciated weakness

of AI is that it tends to do poorly when

it's also perform on new types of data

that's different than the data it has

seen in your data set let me explain

with an example say you built a

supervised learning system that uses a

to be to learn to diagnose pneumonia

from images like these these are you

know pretty high-quality chest x-ray

images but now let's say you take this

AI system and apply it at a different

Hospital or a different Medical Center

where maybe the x-ray technician somehow

strangely had the patients always lied

an angle or sometimes there are these

defects not sure you can see the little

scratches in the image these law other

objects lying on top of the patient's if

the AI system has learned from data like

that on your left maybe taken from a

high quality Medical Center and you take

this AI system and apply it to a

different Medical Center that generates

images like those on the right then this

performance would be quite poor as well

a good AI team would be able to

ameliorate or to reduce some of these

problems

but doing this is not that easy and this

is one of the things that AI is actually

much weaker than humans

if a human has learned from images on

the left they're much more likely to be

able to adapt to images like those on

the right as they figure out that the

patient's just lying on an ankle but an

AI system can be much less robust than

human doctors in generalizing or

freaking out what to do with new types

of data like this I hope these examples

are helping you hone your intuitions

about what a I can and cannot do in case

the boundary between what they can and

cannot do so since fuzzy to you don't

worry that's completely normal

completely okay

in fact even today I still can't open a

project and immediately tell is

something that's peaceful or not and I

often still need reeks of small numbers

of weeks of technical diligence before

forming strong conviction about whether

something is feasible or not but I hope

that these examples can at least help

you start imagining some things in your

company that might be feasible and might

be worth exploring more the next two

videos after this are optional and are a

non-technical description of whether

neural networks and what is deep

learning please feel free to watch those

and then next week will go much more

deeply into the process of what building

and AI project would look like look

forward to see you next week

the terms deep learning and neural

network are used almost interchangeably

in AI and even though they're great for

machine learning there's also been a bit

of hype and bit of mystique about them

this video will demystify deep learning

so that you have a sense of what deep

learning and neural networks really are

let's use an example from demand

prediction let's say you run a website

that sells t-shirts and you want to know

based on how you price the t-shirts

how many units you expect to sell how

many t-shirts you expect to sell you

might then create a data set like this

where the higher the price of the

t-shirt the Lord to demand so you might

fit a straight line to this data showing

that as the price goes up the demand

goes down now demand can never go below

zero so maybe you say that the demand

will flatten out at zero and beyond a

certain point you expect you know pretty

much no one to buy any t-shirts it turns

out this blue line is maybe the simplest

possible neural network you have as

input the price ate and you wanted to

output the estimated demand B so the way

you would draw this as a new network is

that the price would be input to this

little round thing there and this little

round thing outputs yes me to demand in

these terminology of AI this little

round thing here is called a neuron or

sometimes it's called an artificial

neuron and oh it does is compute this

blue curve that I've drawn here on the

left this is maybe the simplest possible

neural network with a single artificial

neuron that just inputs the price and

outputs the estimated demand if you

think of this orange circle this

artificial neuron as a little Lego break

all the other neural network is is if

you take a lot of these Lego breaks and

stack them on top of each other until

you get a big how or a big network

of this Niraj let's look at a more

complex example suppose that instead of

knowing only the price of the t-shirts

you also have the shipping costs that

the customers will have to pay to get

the t-shirts maybe you spend more or

less on marketing in a given week and

you can also make the t-shirt out of a

thick heavy expensive cotton or a much

cheaper more lightweight material these

are some of the factors that you think

will affect the demand for your t-shirts

let's see what a more complex neural

network might look like you know that

your customers care a lot about

affordability so let's say you have one

neuron and let me draw this one in blue

whose job it is to estimate the

affordability of the t-shirts and

because affordability and so

affordability is mainly a function of

the price of the shirt and of the

shipping costs a second thing they'll

affect the demand for your teachers is

awareness

how much are consumers aware that you're

selling this t-shirt so the main thing

that affects awareness is going to be

your marketing so let me draw here a

second artificial neuron that inputs

your marketing budget how much you spent

on marketing and outputs

how aware are consumers of your t-shirt

finally the perceived quality of your

product will also affect demand and

perceived quality would be affected by

marketing if the marketing tries to

convince people this is a high quality

t-shirt and sometimes the price of

something also affects perceived quality

so I'm going to draw here a third

artificial neuron that inputs price

marketing and material and tries to

estimate the perceived quality

of your t-shirts finally now that the

earlier neurons these three blue neurons

have figured out how affordable how much

consumer awareness and what's a

perceived quality you can then have one

more neuron over here that takes us

input these three factors and outputs

the estimated demand so this is a neural

network and its job is to learn to map

from these four inputs that's the input

a to the output B to demand so it learns

this input output or a to be mapping

this is a fairly small neural network

with just four artificial neurons in

practice neural networks used today are

much larger more easily

thousands tens of thousands or even much

larger than that numbers of neurons now

there's just one final detail of this

description that I want to clean up

which is that in the way of describing

neural network

it was as if you had to figure out that

the key factors are affordability

awareness and perceived quality one of

the wonderful things about using neural

networks is that to train a neural

network in other words to build a

machine learning system using a neural

network all you have to do is give it

the input a and the upper B and it

figures out all of the things in the

middle by yourself so to build a neural

network what you would do is feed it

lots of data with the input a and have a

neural network that just looks like this

with a few blue neurons feeding to a

yellow output neuron and then you have

two given data with the demand B as well

and it's the software's job to figure

out what these blue neurons should be

computing so that it can completely

automatically learn the most accurate

possible function mapping from the input

to the output B and it turns out that if

you give this enough data and train a

neural network that is big enough there

can do an incredibly good job mapping

from inputs a to uppers beam so that's a

neural network is a group of artificial

neurons each of which computes a

relatively simple function but when you

stack enough of them together like Lego

breaks they can compute incredibly

complicated functions that give you very

accurate mappings from the input a to

the output B now in this video you saw

an example of neural networks apply to

demand prediction let's go onto the next

video to see a more complex example of

new networks apply to face recognition

in the last video you saw how a neural

network can be applied to demand

prediction but how can a new network

look at the picture and figure out

what's in the picture or listen to an

audio clip and understand what is said

in an audio clip let's take a look at a

more complex example of applying a

neural network to face recognition say

you want to build a system to recognize

people from pictures how can a piece of

software look at this picture and figure

out the identity of the person in it

let's zoom into a little square like

that to better understand how a computer

sees pictures where you and I see a

human eye a computer is that sees that

it sees this grid of pixel brightness

values that tells it for each of the

pixels in the image how bright is that

pixel if it were a black and white or a

grayscale image then each pixel would

correspond to a single number telling

you how bright is that pixel if is a

color image then each pixel would

actually have three numbers

corresponding to how bright are the red

green and blue elements of that pixel so

the new networks job is to take us input

a lot of numbers like these and tell you

the name of the person in the picture in

the last video you saw how a new network

can take as input four numbers

corresponding to the price shipping cost

amount of marketing and cloth material

of a t-shirt and output demand in this

example the neural network just has to

input a lot more numbers corresponding

to all of the pixel brightness values of

this picture if the resolution of this

picture is 1000 pixels by 1000 pixels

then that's a million pixels so if it

were a black and white or grayscale

image this neural network was take as

input a million numbers corresponding to

the brightness of all 1 million pixels

in this image or it was a color image it

would take as input 3 million numbers

corresponding to the red green and blue

values of each of these 1 million pixels

in this image similar to before you will

have many many of these artificial

neurons computing various values and

it's not your job to figure out what

these neurons should compute the new

network will figure it out by itself but

typically when you give it an image the

neurons in the earlier parts of the

neural network will learn to detect

edges in pictures and then little bit

later or learn to detect parts of

objects so may learn to detect eyes and

noses and the shape of cheeks in the

shape of models and then the later

neurons further to the right will learn

to detect different shapes of faces and

it will finally put all this together to

output the identity of the person indeed

image and again part of the magic of

neural networks is that you don't really

need to worry about what it is doing in

the middle all you need to do is given a

lot of data of pictures like this a as

well as of the correct identity B and

the learning algorithm will figure out

by itself

what each of these neurons in the middle

should be computing congratulations on

finishing all the videos for this week

you now know how machine learning and

data signs work I look forward to seeing

you in next week's videos as well where

you learn how to build your own machine

learning or data science project see you

next week

machine learning algorithms can learn

input the output or a to be mappings so

how do you build a machine learning

project in this video you learn what is

the workflow of machine learning

projects let's take a look as a running

example I'm going to use speech

recognition so some of you may have an

Amazon echo or a Google home or Apple

sory device or a Baidu to iOS device in

your homes some years back I done some

work on Google's speech recognition

system and it also led by juice doer oh

s project and today I actually have a

Amazon echo in my kitchen so every time

I'm balding an egg I will say Alexa set

timer for three minutes and then it lets

you know when the Freedmen's are up and

my eggs are ready so how do you build a

speech recognition system that can

recognize when you say Alexa or hey

Google or hey Siri or hello Baidu let's

go through the key steps of a machine

learning project and just for simplicity

I'm going to use Amazon echo or

detecting the Alexa keyword as this

running example if you want to build an

AI system or build a machine learning

system to figure out when a user has

said the word Alexa the first step is to

collect data so that means you go around

and get some people to say the words

Alexa for you and you record the audio

of that and you'll also get a bunch of

people to say other words like hello or

say lots of other words and record the

audio of that as well having collected a

lot of audio data a lot of these audio

clips that people saying either Alexa or

saying other things step two is to then

train the model and this means you will

use a machine learning algorithm to

learn an input to output or a to be

mapping where the input a would be an

audio clip and in the case of the first

audio clip above hopefully it will tell

you that the user said Alexa and in the

case of audio clip two shown on the

right hopefully the system will learn to

recognize that the user has said hello

whenever an AI team starts to train the

model meaning to learn the eight

your input output mapping what happens

pretty much every time is the first

attempt doesn't work well and so

invariably the team will need to try

many times or in a I recall this iterate

many times you have to iterate many

times until hopefully the model looks

like is good enough the third step is to

then actually deploy the model and what

that means is you put this AI software

into an actual small speaker and ship it

to either a small group of test users or

to a large group of users what happens

in a lot of AI products is that when you

ship it you see that it starts getting

new data and it may not work as well as

you had initially hoped so for example I

am from the UK so I'm going to pick on

the British but let's say you had

trained your speech recognition system

on American accented speakers and you

then ship this small speaker to the UK

and you start having British accent -

people say Alexa then you may find that

it doesn't recognize their speech as

well as you had hoped and when that

happens hopefully you can get data back

of cases such as maybe British accented

speakers was not working as well as

you're hoping and then use this data to

maintain and to update the model so to

summarize the key steps of a machine

learning project are to collect data to

train the model that a to be mapping and

then to deploy the model and throughout

these steps there is often a lot of

iteration meaning fine tuning or

adapting the model to work better or

getting data back even after you've

shifted to hopefully make the product

better which may or may not be possible

depending on whether you're able to get

data back let's look at these three

steps and see how they apply on the

different projects on building a key

component of a self-driving car so

remember the key steps that collect data

at Raymond and deploy model since we

revisit these three steps on the next

slide let's say you're building a

self-driving car one of the key

components is a self-driving car is a

machine learning algorithm that takes as

input say a picture of what's in front

of your car and tells you where are the

other calls so what's the first step of

building this machine

learning system hopefully you remember

from the last night that the first step

was to collect data so if you go is have

a machine learning algorithm they could

take us input an image and output the

position of other cause the data you

need to collect would be both images as

well as position of other costs that you

want to a our system to output so let's

say you start off with a few pictures

like this these are the inputs a to the

machine learning algorithm you need to

also tell it what is the output B you

would want and so for each of these

pictures you would draw a rectangle

around the cause in the picture that you

wanted to detect and on this line I'm

hand drawing these rectangles but in

practice you will use some software that

lets you draw perfect rectangles rather

than these hand-drawn ones and then

having created this data set what was

the second step hope you remember that

the second step was to train them

although now invariably when you're AI

engineers start training a model they'll

find initially that it doesn't work that

well for example given this picture

maybe the software the first few tries

things that that is a car and it's only

by iterating many times that you

hopefully get a better result like

figuring out that that is where the car

actually is finally what was the third

step it was to deploy the model of

course in the self-driving world is

important to treat safety as number one

and deploy model or to test the model

only in ways they can preserve safety

but when you put the software in cars on

the road you may find that there are new

types of vehicles say golf cause that

the software isn't detecting very well

and so you get data back say pictures of

these golf cars use the new data to

maintain and update the model so that

hopefully you can have your AI software

continually get better and better to the

point where you end up with a software

that can do a pretty good job detecting

other costs from pictures like these in

this video you learn what are the key

steps of a machine

project which ought to collect data to

train them although and then to deploy

the model NYX let's take a look at

whether the key steps or what does it

work though of a data science project

let's go onto the next video

unlike a machine learning project the

output of a data science project is

often a set of actionable insights of

insights that may cause you to do things

differently so data science projects

have a different workflow the machine

learning projects let's take a look at

one of the steps of a data science

project as our running example let's say

you want to optimize a sales funnel say

you run a ecommerce or online shopping

website that sells coffee mugs and so

for a user to buy a coffee mug from you

there's a sequence of steps they'll

usually follow first they'll visit your

website and take a look at the different

coffee mugs on offer then eventually if

you get to a product page and then

they'll have to put it into the shopping

cart and go to the shopping cart page

and then they'll finally have to

checkout so if you want to optimize the

sales funnel to make sure that as many

people as possible get through all of

these steps how can you use data signs

to help with this problem let's look at

the key steps of a data science project

the first step is to collect data so on

a website like the one we saw you may

have a data set that's forced when

different users go to different web

pages in this simple example I'm

assuming that you can figure out the

country that the users are coming from

for example by looking at their

computers address called an IP address

and figuring out what is the country

from which they're originating but in

practice you can usually get quite a bit

more data about users than just what

country they're from the second step is

to then analyze the data your data

science team may have a lot of ideas

about what is affecting the performance

of your sales funnel for example they

may think that overseas customers are

scared off by the International shipping

costs which is why a lot of people go to

the checkout page but don't actually

check out and if that's true then you

might think about whether to put part of

shipping costs into the actual product

costs or your data science team may

think there are blips in the data

whenever there's a holiday maybe more

people will shop around the holidays

because the bank gives or maybe fewer

people will shop around the holidays

because they're staying home rather than

you know sometimes shopping from their

work computers and in some countries

there may be time of day blips where in

countries that observe a siesta so a

time of rest like an afternoon rest

there may be fewer shoppers online and

so your sales may go down and they might

didn't suggest that you should spend

fewer advertising dollars during the

period of CES there because fewer people

will go online to buy at that time so a

good data science team may have many

ideas and so they try many ideas or will

say anyway many times to get good

insights finally the data science team

will destroy these insights down to a

smaller number of hypotheses about ideas

of what could be going wrong what could

be going poorly as well as a smaller

number of suggested actions such as

incorporating shipping costs into the

product cost rather than having it as a

separate line item when you take some of

these suggested actions and deploy these

changes to your website you then start

to get new data back as users behave

differently now that you advertise

differently at the time of Siesta or of

a different checkout policy and then

your data science team can continue to

collect data and reanalyze the new data

periodically so see if they can come up

with even better hypotheses or even

better actions over time so the key

steps of a data science project are to

collect the data to analyze the data and

then to suggest hypotheses and actions

and then to continue to get the data

back and we analyze the data

periodically let's take this framework

and apply it to a new problem to

optimizing a manufacturing line and so

we'll take these three steps and use

them on the next slide as well let's say

you run the factories as manufacturing

thousands of coffee mugs a month for

sale and you want to optimize the

manufacturing line so these are the key

steps in manufacturing coffee maps step

one is the mixer clay so make sure the

appropriate amount of water is added

step two is to take the screen and to

shape the mugs then you have to add the

glaze so that the coloring protective

cover then you have to heat this mug and

we call that firing the kiln and finally

you would inspect the

to make sure there aren't dents in the

mug and it isn't cracked before you

should put a customer's so a common

problem in manufacturing is to optimize

the yield of this manufacturing line to

make sure that as you damage coffee mugs

get produced as possible because those

are coffee rust you have to throw away

resulting in time and material waste

what's the first step of a data science

project I hope you remember from the

last slide - the first step is to

collect data so for example you may save

data about the different batches of clay

that you've mix such as who supply the

clay and how long did you mix it or

maybe how much moisture wasn't it clear

how much water did you add you might

also collect data about the different

batches of mugs you made so how much

humidity wasn't that badge what was the

temperature in the kiln and how long did

you fire it in the kiln given all this

data you would then ask the data science

team to analyze the data and they would

ask before it read many times to get

good insights and so they might find out

for example that whenever the humidity

is too low and the kiln temperature is

too hot that it cracks in the mug or

they may find out that because it's

warmer in the afternoon that you need to

adjust the humidity and temperature

depending on the time of day based on

the insights from your data science team

you get suggestions for hypotheses and

actions on how to change the operations

of the manufacturing line in order to

improve the productivity of the line and

when you deploy the changes you then get

new data back that you can be analyzed

periodically so you can keep on

optimizing the performance of your

manufacturing line to summarize the key

steps of a data science project are to

collect the data to analyze the data and

then to suggest hypotheses and actions

in this video in the last video you saw

some examples of machine learning

projects and data science projects it

turns out that machine learning and data

science are affecting almost every

single job function what I want to do in

the next video is show you how these

ideas are affecting many job functions

including perhaps yours and certainly

that many of your colleagues let's go

into the next video

you

data is transforming many different job

functions whether you work in recruiting

or sales or marketing or manufacturing

or agriculture data is probably

transforming your job function what's

happened in the last few decades is the

digitization of our society so rather

than handing out paper surveys like

these surveys are more likely to be done

in digital format or doctors still write

some handwritten notes but the doctors

handwritten note is increasingly likely

to be a digital record and so to this in

just about every single job function and

this availability of data means that

there's a good chance that your job

function can be helped with tools like

data science or machine learning let's

take a look and in this video I want to

run through many different job functions

and discuss how data science and machine

learning can or will impact these

different types of jobs there's lots of

sales you've already seen in the last

video how data science can be used to

optimize a sales funnel

how about machine learning if you're a

salesperson you may have a set of leaves

about different people that you can

reach out to to convince them to buy

something from your company machine

learning can help you prioritize these

leads so you might want to prioritize

calling out the CEO of the large company

rather than the intern at a much smaller

company and this type of automated lead

sorting is making salespeople more

efficient let's look at more examples

let's say your manufacturing line

manager you've already seen how data

science can help you optimize a

manufacturing line

how about machine learning one of the

steps of this manufacturing process is

the final inspection and in fact today

in many factories there can be hundreds

or thousands of people using the human

eye to check over objects maybe coffee

mugs maybe other things to see if

there's scratches or dents and that's

called inspection so machine learning

can take us input data set like this and

learn to automatically figure out if a

coffee mug is defective or not and by

automatically finding scratches or dents

it can

reduce labor costs and also improve

quality in your factory this type of

automated visual inspection is one of

the technologies that I think will have

a big impact on manufacturing this is

something I've been working on myself as

well let's see more examples how about

recruiting when recruiting someone to

join your company there may be a pretty

predictable sequence of steps where your

recruiter or someone else would send an

email to a candidate and then you'd have

a phone call of them bring them on-site

for an interview and then extend an

offer and maybe close the offer similar

to how data science can be used to

optimize a sales funnel recruiting can

also use data science to optimize a

recruiting funnel and in fact many

recruiting organizations are doing so

today for example if you find that

hardly anyone is making it from the

phone screen step to the on-site

interview step then you may conclude

that maybe too many people are getting

to the phone screen stage or maybe the

people doing the phone screen are just

being too tough and they should let more

people get to the on-site interview

stage this type of data science is

already having an impact on recruiting

what about machine learning projects

well one of the steps of recruiting is

to screen a lot of resumes to decide who

to reach out to so you may have to look

at my resume and says yes let's email

them don't get a different one to say no

let's not move ahead with this candidate

and machine learning is starting to make

us weigh into automated resume screening

this does raise important ethical

questions such as making sure that your

AI software does not exhibit undesirable

forms of bias and treat people fairly

but machine learning is starting to make

inroads into this and hope can do so

while making sure that the systems are

ethical and fair in the final week of

this AI for everyone calls you also

learn more about the issues of fairness

and ethics in AI wonder if you work in

marketing one of the common ways to

optimize that performance on the website

is called a be testing in which you

launch two versions of website here in

version 8 has a red button version B has

a green button and you'd measure which

website causes people to click through

so with this type of data a data science

team can help you gain insights and

suggest hypotheses or actions for

optimizing a website how about machine

learning and marketing today a lot of

websites will give customized product

recommendations to show you the things

you are most likely to want to buy and

there's actually significant increases

sales on these websites for example a

clothing website after has seen the way

I shop after a while will hopefully just

recommend blue shirts to me because

that's frankly pretty much the only type

of shirt I ever buy but maybe other

customers will have more diverse and

more interesting recommendations than

mine but today these customized product

recommendations actually drive a large

percentage of sales on many large online

e-commerce websites one last example

from a totally different sector let's

say you work in agriculture maybe you're

a farmer working on the large industrial

farm how can data science help you today

farmers already using data signs for

crop analytics where you can take data

on the soil conditions the weather

conditions the prices of different crops

in the market and have data science

teams make recommendations to what to

plant when to plant so as to improve use

while maintaining the condition of the

soil on your farm

this type of data science is and will

play a bigger and bigger role in

agriculture let's also look at the

machine learning example I think one of

the most exciting changes to agriculture

is precision agriculture here's a

picture that I took on a farm with my

cell phone on the upper right is a

cotton plant and shown in middle is a

weed and so with machine learning we're

starting to see products that can go on

to the farms take a picture like this

and spray a read color in a very precise

way just onto the weeds so that it gets

your the read but without having to

spray an excessive amount of read colors

this type of machine learning technology

is both helping farmers increased crop

use while also hoping to preserve the

environment in this video you saw how

all of these job functions everything

from sales recruiting to marketing to

manufacturing to farming agriculture how

all of these job functions are being

affected

data by data science and the machine

learning it seems like there's a lot of

different things you could do with AI

but how do you actually select a

promising project to work on let's talk

about that in the next video

if you want to try your hand at an AI

project how do you select a worthwhile

project to work on don't expect an idea

it's an S we come overnight sometimes it

happens but sometimes it also takes a

few days or maybe a few weeks to come up

with a worthy idea to pursue in this

video you see a framework for

brainstorming potentially exciting the

AI projects to pursue let's say you want

to build an AI project for your business

you've already seen that AI can't do

everything and so there's going to be a

certain set of things that is what AI

can do so let's let the circle represent

a set of things that AI can do now

there's also going to be a certain set

of things that is valuable for your

business so let's let this second circle

represent a set of things that are

valuable for your business what you

would like to do is try to select

projects that are at the intersection of

these two sets so you select projects

hopefully that are both feasible that

can be done with AI and that are also

valuable for your business so AI experts

will tend to have a good sense of what

is and what isn't in the set on the left

and domain expense expense in your

business be it sales or marketing or

agriculture or something else what have

a best sense of what is actually

valuable for your business so when

brainstorming projects that AI can do

and are valid for your business I will

often bring together a team comprising

both people knowledgeable AI as well as

experts in your business area to

brainstorm together so that together

they can try to identify projects at the

intersection of both of these two sets

so sometimes we also call these

cross-functional teams and that just

means a team that includes both AI

experts as well as domain experts

meaning experts in your area of business

when brainstorming projects there's a

framework that I've used with a lot of

companies that are found to be useful so

let me share with you three principles

or three ideas for how you can have a

team brainstorm projects first even

though there's been a lot of press

coverage about AI automating jobs away

this is an important societal issue that

needs to be addressed when thinking

about concrete AI projects I find it

much more useful to think about

automating tasks rather than automating

jobs take call center operations there

longer tasks that happen in the call

center ranging from people pick up the

phone to answering phone calls to

replying to emails to taking specific

actions such as issuing a refund on

behalf of a customer requests but along

with these tasks that employees in the

call center do there may be one call

routing or email routing that may be

particularly amenable so machine

learning automation and it's been

looking at all these tasks that the

group of employees do and selecting one

that we allow you to select the most

fruitful project for automation in the

near term let's look at another example

the job of radiologists there's no

longer press about how a I'm a automate

radiologist jobs but radiologists

actually do a lot of things they read

x-rays that's really important but they

also engage in their own continuing

education they consulted other doctors

they may mentor younger doctors some of

them also consult directly with patients

and so it's by looking at all of these

tasks that the radiologist does that you

may identify one of them let's say AI

assistants or AI automation for reading

x-rays that allows you to select the

most fruitful projects to work on so

whether we recommend is if you look in

your business

think about the tasks that people do to

see if you can identify just one of them

or just a couple of them that may be

automatable using machine learning when

our meeting CEOs of large companies to

brainstorm AI projects for the company a

common question out also ask is what are

the main drivers of business value and

sometimes finding AI solution so they

design solutions to augmentis can be

very valuable finally a third question

that I've asked there's sometimes letter

valuable project ideas is what the main

pain points in your business some of

them could be soft of AI some of them

can't be soft for the eye but by

understanding the main pain points of

the business that can create a useful

starting point for brainstorming AI

projects as well I have one last piece

of advice for brainstorming AI projects

which is that you can make progress even

without big data

even without tons of data now don't get

me wrong having more data almost never

hurts other than maybe needing to pay a

bit more for disk space or network

bandwidth to transmit and store the data

having more data almost always is only

helpful and I love having lots of data

it is also true that data makes some

businesses like web search defensible

web search is a longtail business

meaning that there are a lot of very

very rare web queries and so seeing what

people click on when they search on all

of these rare web queries does help the

leading web search engines have a much

better search experience so big data is

great when you can get it but I think

big data is also sometimes overhyped and

even with a small data set you can still

often make progress here's an example

let's say you're building a automated

visual inspection system for the coffee

mug so you want to automatically detect

that the coffee mug on the right is

defective well if you had a million

pictures of good coffee mugs and

defective coffee mugs it'd be great to

have that many examples of pictures of

good and bad coffee mouths to feed

through AI system but I hope that you

have not manufactured a million

defective coffee mouths because that

feels like a very expensive thing to

have to throw away so sometimes with as

few as a hundred pictures or maybe a

thousand or sometimes maybe as few as

ten you may be able to get started on a

machine learning project the amount of

data you need is very problem dependent

and speaking with a ai engineer on AI

expert would help you get better since

there are some problems for a thousand

images may not be enough where you do

need big data to get good performance

but my advice is don't give up just

because you don't have a lot of data to

start off with and you can often still

make progress even with a small data set

in this video you saw a brainstorming

framework and a set of criteria for

trying to come up with projects that

hopefully can be doable with AI and

they're also valuable for your business

now having brainstormed elicit projects

how do you select one or select a small

handful to actually commit to and work

on let's talk about that in the next

video

maybe have a lot of ideas for possible

AI projects to work on

but before committing to one how do you

make sure that this really is a

worthwhile project if is a quick project

that might take you just a few days

maybe just jump in right away and see

the worse or not but some AI projects

may take many months to execute in this

video I want to step you through the

process that I use to double check if a

project is worth that many months of

effort let's take a look before

committing to a big AI project I will

usually conduct due diligence on it due

diligence has a specific meaning in the

legal world but informally it just means

that you want to spend some time to make

sure what your hope is true really is

true you've already seen how the best AI

projects are ones that are feasible so

it's something that a I can do as well

as valuable we really want to choose

projects to the at the intersection of

these two sets so to make sure a project

is feasible I will usually go through

technical diligence and to make sure

that the project is valuable I will

usually go through a business diligence

process let me tell you more about these

two steps technical diligence is the

process of making sure that the AI

system you hope to build really is

doable really is feasible so you might

talk to AI X phase about whether or not

the AI system can actually meet the

desired level of performance for example

if you are hoping to build a speech

system that is 95% accurate Consulting

of AI experience or perhaps reading some

of the trade literature can give you a

sense of whether this is doable or not

or if you want a system to inspect

coffee mugs in the factory and you need

your system to be 99% accurate again is

this actually doable with today's

technology a second important question

for technical diligence is how much data

is needed to get to this design level

performance and do you have a way to get

that much

third would be engineering timeline to

try to figure out how long it will take

and how many people will take to build

the system that you would like to have

built in addition to technical divisions

I will often also conduct business

diligence to make sure that the project

you envision really is valuable for the

business so a lot of AI projects would

drive value through lowering costs for

example by automating a few tasks or by

squeezing more efficiency out of a

system a lot of AI systems can also

increase revenue for example driving

more people to check out in your

shopping cart or you may be building an

AI system to help you launch a new

product or a new line of business so

business diligence is the process of

thinking through carefully for the AI

system that you're building such as a

speech recognition system that's 95%

accurate or a visual inspection system

does 99.9% accurate would allow you to

achieve your business goals whether your

business goal is to improve your current

business or to even create brand new

businesses in your company when

conducting business diligence I'll often

end up building spreadsheet financial

models to estimate the value

quantitatively such as estimate how many

dollars are actually saved or what do we

think is a reasonable assumption in

terms of increase revenue and to model

out the economics associated with a

project before committing to many months

of effort on the project although not

explicitly listed on this slide

one thing I hope you're also considering

is the third type of diligence which is

ethical diligence I think there are a

lot of things that AI can do that will

even make a lot of money but that may

not make society better off so in

addition to technical diligence and

business diligence I hope you also

conduct ethical diligence and make sure

that what are you doing is actually

making humanity and making society

better off we also talked more about

this in the last week of this course as

well as you're planning out your AI

project you also have to decide do you

want to

or by this is an age-old question in the

IT world and we're facing this question

in AI as well for example hardly any

companies built their own computers

these days they buy someone else's

computers and hardly any companies build

their own Wi-Fi routers just by a

commercial Wi-Fi router

how about machine learning and data

signs machine learning projects can be

in-house or outsourced I've seen both of

these models used successfully sometimes

if you outsource and machine learning

project you can have access much more

quickly to talent and get going faster

on the project it is nice if eventually

you build your own in-house AI team and

can also do these projects in-house you

hear more about this when we talk about

the AI translation playbook in greater

detail next week unlike machine learning

projects though data science projects

are more commonly done in-house they're

not impossible to outsource you can

sometimes outsource them but what I've

seen is that data science projects are

often so closely tied to your business

then it takes very deep day-to-day

knowledge about your business to do the

best data science projects and so just

as a percentage as a fraction

I see data science projects in house

more than machine learning projects

finally in every industry some things

will be industry standard and you should

avoid building those a common answer to

the build versus buy question was don't

the things they're going to be quite

specialized to you or completely

specialized to you or they'll allow you

to build a unique defensive advantage

but the things that will be industry

standard probably some other company

will build and it'll be more efficient

for you to just buy it rather than

building in hosts one of my team's had a

really poetic phrase which is don't

sprint in front of a train and what that

means is if this is a train running on

the railway tracks

and that's the nor chimney with the puff

of smoke what you don't want to do is to

be the person or the engineer trying to

sprint faster and faster ahead of the

Train the Train is the industry standard

solution and so if there's a company

maybe a Santa maybe a big company or

maybe an open-source effort that is

building an industry standard solution

then you may want to avoid trying to run

faster and faster to keep ahead of the

Train

because even though you could sprint

faster in the short term eventually the

train will catch up and you know crush

someone trying to sprint in front of a

train so when there's a massive force of

an industry standard solution that is

being built you might be better off just

embracing an industry standard or

embracing someone else's platform rather

than trying to do everything in-house

we are live in a world of limited

resources limited time limited data

limited and drain resources and so I

hope you can focus those resources on

the project so that most unique can make

the biggest difference to your company

through the process of technical

diligence as well as business diligence

I hope you can start to identify

projects that are potentially valuable

or that seem promising for your business

if the project is a big component maybe

I'll take many months to do it's not

unusual for me to spend even a few weeks

conducting this type of diligence before

committing to a project now say you've

found a few promising projects how do

you engage from an AI team how do you

work um in the I team to try to get

these projects done let's talk about

that in the next video

you

so you found

exciting project that you want to try to

excuse on how do you work of an AI team

on this project in this video you

learned how a I teams think about data

and therefore how you can interact with

AI teams to help them succeed on a

project now there is one caveat which is

what if you have a cool idea but you

don't have access to an AI team you

don't have any access to any AI

engineers

fortunately in today's world if either

you yourself or you can encourage some

of your injuring your friends to take an

online course or two on machine learning

or deep learning that often will give

them enough knowledge to get going and

make a start of an attempt make a

reasonable attempt on these types of

projects so let's talk about how you can

work with an AI team first it really

helps your AI team if you can specify an

acceptance criteria for the project I've

done a lot of work in automated visual

inspection so I'm going to use that as a

running example in these few slides

let's say your goal is to detect defects

in coffee mugs with at least 95 percent

accuracy so that can be your acceptance

criteria for this project but 95 percent

accuracy how do you measure accuracy one

of the things that the AI team would

need is a data set on which to measure

their accuracy so data set is just a set

of pictures like these together with the

labels with the design output be that

the first two coffee mugs are okay and

the third one is defective so as part of

your specification for the acceptance

criteria you should make sure that the

AI team has a data set on which to

measure the performance so that they can

know if they've achieved 95 percent

accuracy the formal term for this data

set is called a test set and the test

set may not need to be too big maybe a

thousand pictures will be just fine for

this example but if you consulted that

AI expert they can give you a better

sense of how big the test set needs to

be for them to be able to evaluate

whether or not they're getting to 95%

accuracy one novel part of a

systems is that the performance is

usually specified in a statistical way

so rather than a free friend AI system

that just does something perfectly you

see very often that we want any AI

system that performs at a certain

percentage accuracy like this example

here so when specifying your acceptance

criteria think of whether your

acceptance criteria needs to be

specified in a statistical way where you

specify on average our does or what

percent of time it has to get the right

answer let's dive more deeply into the

concept of a test set this is how AI

teams think about data AI teams group

data into two main data says the first

called the training set and the second

called the test set which we've already

talked a bit about the training set is

just a set of pictures together with

labels showing whether each of these

pictures is of a coffee mug that is okay

or defective so the training set gives

examples of both the input a the

pictures of the coffee mouse as well as

the desired output B whether it's okay

or defective and so given this training

set what a machine learning algorithm

will do is learn in other words compute

or figure out some mapping from A to B

so that you now have a piece of software

they can take as input the input a and

try to figure out what is the

appropriate output B so the training set

is the input to the machine learning

software that lets it figure out what is

this a to b mapping the second data said

that an AI team will use is the test set

and as you've seen this is just another

set of images that's different from the

training set also what the labels

provided the way an AI team will

evaluate their learning algorithms

performance is to give the images in the

test set to the AI software and see what

the AI software outputs for example if

on these three tests set images the AI

software outputs okay for this okay for

this and also okay for this then we will

say that they got two out of three

examples right

so that's a 66.7%

accuracy in this figure the training set

and test sets are both only three

pictures in practice both of these data

sets would be much bigger of course and

you find it for most problems the

training set is much much much bigger

than the test set but you can talk to AI

engineers to find out how much data they

need for a given problem

finally for technical reasons some AI

teams will need not just one but two

different test sets if you hear AI teams

talk about development or dev or

validation sets that's the second test

set that they're using the reasons why

they need to test says is quite

technical and beyond the scope of this

course but if an AI team asks you for

two different test sets is quite

reasonable to try to provide that to

them before wrapping up this video one

pitfall I want to urge you to avoid is

expecting a hundred percent accuracy

from your AI software here's what I mean

let's say this is your test set which

you've already seen on the last slide

but let me add a few more examples to

this test set here are some of the

reasons it may not be possible for a

piece of AI software to be a hundred

percent accurate first machine learning

technology today despite being very

powerful slow has limitations and they

just can't do everything and so you may

be working on a problem that is just

very difficult even for today's machine

learning technology second insufficient

data if you don't have enough data

specifically if you don't have enough

training data for the AI software to

learn from it may be very difficult to

get a very high level of accuracy third

data is messy and sometimes data can be

mislabeled

for example this green coffee mug here

looks perfectly okay to me so the label

of it being a defect looks like an

incorrect label and that would hurt the

performance of your AI software and data

can also be ambiguous for example it

looks like this coffee mug has a small

scratch over there and it's a pretty

small scratch so maybe we will think of

this though

hey that maybe this should actually have

been a defect or maybe even different

experts won't agree if this book of

coffee mug is okay I should pass the

inspection step some of these problems

can be ameliorated for example if you

don't have enough data maybe you can try

to collect more data and more data more

often help or you can also try to clean

up mislabeled data or try to get your

factories expensed come to better

agreement about these ambiguous labels

so there are ways to try to make these

things better but a lot of AI systems

are incredibly valuable even without

achieving a hundred percent accuracy so

I would urge you to discuss with your AI

engineers what is a reasonable level of

accuracy to try to accomplish and then

try to find something that passes both

technical diligence as well as business

diligence without necessarily needing a

hundred percent accuracy congratulations

on finishing all the videos for this

week you now know what it feels like and

what it takes to build an AI project and

I hope you start brainstorming and

exploring some ideas there is one more

optional video describing some of the

technical tools that AI teams use they

can watch if you wish but either way I

look forward to seeing you next week

where you learn how a AI projects fit in

the context of a bigger company look

forward to seeing you next week

when you work with AI teams you may hear

them refer to the tools that they're

using to build these AI systems in this

video I want to share of you some

details and names of the most commonly

used AI tools so that you people better

understand what these AI engineers are

doing we're fortunate that the AI world

today is very open and many teams will

openly share ideas of each other there

are great machine learning open source

frameworks that many teams are used to

build their systems so if you hear of

any of these 10sec philip I don't care

is MX nets the antique a cafe paddle

paddle so I can learn our or Wecker all

of these are open source machine

learning frameworks that help AI teams

be much more efficient in terms of

writing software along of AI technology

breakthroughs are also published freely

on the internet on this website called

archive it's felt like this I hope that

other academic communities also freely

share their research since I've seen

firsthand how much does accelerates

progress in the whole field of AI

finally many teams will also share their

code freely on the internet most

commonly on the website called github

this has become the de facto repository

for open-source software in AI and in

other sectors in AI and by using

appropriately licensed open-source

software many teams can get going much

faster than if they had to build

everything from scratch so for example

if I search online for face recognition

software on github you might find a web

page like this and if you scroll down

this actually has a pretty good very

readable description of software that is

made available on this website for

recognizing people's faces and even

finding parts of people's faces there's

just a ton of software that is freely

downloadable for doing all sorts of

things on the internet and just double

check the license or AI team would

double check the license before using it

of course but a lot of the software is

open source or otherwise very pretty

mystically license for anyone to use

although github is a technical website

built for engineers if you want you

should feel free to play around github

and see what are the types of AI

software people have released online as

well in addition to these open source

technical tools you often also hear AI

engineers talk about CPUs and GPUs

here's what these terms mean a CPU is

the computer processor in your computer

whether is your desktop your laptop or a

computer server off in the cloud CPU

stands for a central processing unit and

CPUs are made by Intel and AMD and a few

other companies this does a lot of the

computation in your computer GPU stands

for graphics processing unit

historically the GPU was made to process

pictures so if you play a video gaem is

probably a GPU that is drawing the fancy

graphics but what we found several years

ago was that the hardware there was

originally built for processing graphics

turns out to be very very powerful for

building very large new networks or very

large deep learning algorithms given the

need to build very large deep learning

or very large neural network systems the

AI community has had this insatiable

hunger for more and more computational

power to train bigger and bigger neural

networks and GPUs have proved to be a

fantastic fit to this type of

computation that we need to have done to

train very large neural networks so

that's why GPUs are playing a big role

in the rise of deep learning and then

there is a company that's been selling

many GPUs but other companies including

Qualcomm as well as Google making his

own TP use are increasingly making

specialized hardware for powering these

very large neural networks finally you

might hear about cloud versus

on-premises or for short on the Prem

deployments cloud deployments refer to

if you rent compute service such as from

Amazon's AWS or Microsoft Azure or

Google's GCP in order to use someone

else's service to do your computation

whereas an on-prem deployment means

buying your own compute service and

running the service locally in your own

company a detailed exploration of the

pros and cons of these two options is

beyond the scope of this video a lot of

the world is moving to cloud deployments

but the research online do you find many

articles talking about the pros and cons

of cloud versus on-prem deployments

there is one last term you might hear

about which is edge deployments if you

are building a self-driving car there's

not enough time to send data from a

self-driving car to a cloud server to

decide if you can stop the car or not

and then send that message back to the

self-driving car so the computation has

to happen usually in the computer right

there inside the car that's called an H

deployment where you put a processor

right where the data is collected so

that you can process the data and make a

decision very quickly

without needing to transmit the data

over the internet to be processed

somewhere else if you look at some of

the small speakers in your home as well

this too is an H deployment where some

not all but some of the speech

recognition toss is done by a processor

that is built-in right there into the

small speaker that is inside your home

the main advantage of a deployment is it

can increase response time of the system

and also reduce the amount of data you

need to send over the network but there

are many pros and cons as well about a

tree versus cloud versus on-prem

deployments that you can also search

online to read more about thanks for

finishing this optional video on the

technical tools that AI engineers use

hopefully when you hear them refer to

some of these tools you start to have a

better sense of what they mean I look

forward to seeing you next week