Artificial intelligence in practice- part-1

Some Impressionistic Take away from the Book of
Bernard Marr & Matt Ward
Artificial Intelligence in Practice
( Part – 1)
( How 50 Successful companies used AI & Machine Learning to Solve problems)
Ramki
ramaddster@gmail.com

The Summary of this book is made in 4 parts
due to large coverage of the book .
This is Part – 1

About the Authors
BERNARD MARR is the founder and CEO of Bernard Marr & Co
and an internationally best-selling business author, futurist,
keynote speaker and strategic advisor to companies and
governments. He is one of the world's most highly respected
voices and a renowned expert when it comes to topics such as
artificial intelligence and big data. Marr advises many of the
world's best-known organizations on strategy, digital
transformation and business performance. He is the author of Big
Data in Practice: How 45 Successful Companies used Big Data
Analytics to Deliver Extraordinary Results and Big Data: Using
SMART Big Data, Analytics and Metrics To Make Better Decisions
and Improve Performance, both published with Wiley.
MATT WARD is the research lead for Bernard Marr & Co. Matt has a background in
investigative journalism and spent the last few years working closely with Bernard Marr
on the latest technology topics. Matt is an expert and experienced writer in the field of
business technology and artificial intelligence, where he has worked with companies
such as IBM, Intel, Citibank and NASA.

 Artificial Intelligence in Practice is a fascinating look into how companies
use AI and machine learning to solve problems. Presenting 50 case
studies of actual situations, this book demonstrates practical applications
to issues faced by businesses around the globe.
 The rapidly evolving field of artificial intelligence has expanded beyond
research labs and computer science departments and made its way into
the mainstream business environment.
 Artificial intelligence and machine learning are cited as the most important
modern business trends to drive success. It is used in areas ranging from
banking and finance to social media and marketing.
 This technology continues to provide innovative solutions to businesses of
all sizes, sectors and industries. This engaging and topical book explores
a wide range of cases illustrating how businesses use AI to boost
performance, drive efficiency, analyze market preferences and many
others.
 Best-selling author and renowned AI expert Bernard Marr reveals how
machine learning technology is transforming the way companies conduct
business.
Prelude

 This detailed examination provides an overview of each company,
describes the specific problem and explains how AI facilitates
resolution. Each case study provides a comprehensive overview,
including some technical details as well as key learning summaries:
 Understand how specific business problems are addressed by
innovative machine learning methods
 Explore how current artificial intelligence applications improve
performance and increase efficiency in various situations
 Expand your knowledge of recent AI advancements in technology
 Gain insight on the future of AI and its increasing role in business
and industry
 Artificial Intelligence in Practice: How 50 Successful Companies
Used Artificial Intelligence to Solve Problems is an insightful and
informative exploration of the transformative power of technology
in 21st century commerce.
Prelude

Some understanding of Technology
A brief guide to some of the most common
and some of the latest terminology being used
when discussing cutting-edge
( It is important to read & understand the
next few slides which gives the flavor of
technology)

Technology
AlphaGo
 It is an AI which became the first computer program to beat a
professional player at the board game Go.
 Game playing has often been a field in which computer scientists
have sought to prove that machines can outperform humans.
 However earlier applications such as chess computers are not
considered “true” AI today because they don’t really learn – they
simply rely on brute force to consider every permutation of a
structured dataset (all of the moves possible in a game of chess.)
 AlphaGo uses deep learning to refine its algorithms based on the
results of historical games, and from running simulated games
against itself.
 This means it can be considered to be learning and comes closer to
what we consider “true” (human-like) intelligence..

Technology
Artificial Intelligence
 This is the original, catch-all term for “machines which can think”,
first conceived by philosophers and storytellers in ancient times.
 Technological advancement has brought them closer to reality
and also redefined what we consider “intelligence” when it
comes to machines.
 Rather than walking, talking automatons, today’s AI’s are more
likely to take the form of discreet computer code dedicated to
handing a particular task in an intelligent way.
Big Data
 The “fuel” of AI. Knowledge unlocks understanding and wisdom.
AI platforms leverage the huge volume, variety and velocity of
information available in today’s digitized world to learn faster and
make increasingly well-informed recommendations and
decisions.

Technology
Cognitive computing
 Cognitive computing is the process by which computers think and learn,
as well as the development of these processes.
 These give rise to artificial intelligence, machine learning, deep learning
and all technologies which involve simulating human thought and decision-
making.
 In practice, the term is often used synonymously with modern, application-
focused AI.
Deep learning
 This is a subfield of machine learning (see below) which uses many layers
of artificial neural networks to handle processing of data in increasingly
complex ways.
 This means that classification (sorting into sets) can be done more
precisely and pattern recognition is more sophisticated.
 These are two of the most useful fundamental tasks that AI carries out
today, meaning Deep Learning is a cutting-edge and very active field of
research. Layers of neural networks stacked on top of each other to be
used in deep learning are known as deep neural networks.

Technology
Generalized AI
 Generalized AI is a concept – widely thought to still be some way off – of a
machine which can carry out any job it is told to do.
 An android such as those seen in Star Trek or Blade Runner – who could be
given a mop and told to clean a floor, or given a weapon and told to defend
against attacking Klingons, would be an archetypal example.
 While advances such as machine learning and deep neural networks point
towards it being something that we will achieve in the future, currently the
majority of AI research focuses on creating applied or specialized AIs (see
below).
Image recognition
 Teaching machines to recognize and classify objects visually – by inputting
visual data – is an important foundation of AI because visual information is so
valuable to humans, and AI seeks to emulate human thought processes.
 Either using cameras or raw image data such as picture or video files,
computers are being taught to classify images according to what they depict,
using pattern recognition to identify key features.
 Advances in machine learning have greatly improved the ability of computers
to do this, as they have become able to teach themselves from vast image
databases, increasing their probability of outputting accurate results.

Technology
Machine learning
 Often used synonymously with AI these days, but there is an
important distinction.
 While AI applies to the entire concept of “thinking” machines from
sci-fi robots to self-learning computer code being developed by
business and academia today, Machine Learning (ML) is the
practical implementation that is generating the biggest
breakthroughs in the real world.
 At its most basic it is technology designed around the principle that
rather than have to teach machines to carry out every task, we
should just be able to feed them data and allow them to work out
the rules by themselves.
 This is done through a process of simulated trial-and-error where
machines crunch datasets through algorithms which are capable of
adapting, based on what they learn from the data, in order to more
efficiently process subsequent data.

Technology
Natural Language Processing
Natural Language Processing (NLP) technology is concerned with
building machines which can understand human speech patterns.
Because spoken communication comes far more naturally to us
than writing computer code, it makes sense that machines, with
their superior processing powers, learn to adapt to us by
understanding and speaking our language, rather than us adapt to
them!
Due to the huge variance in human languages and the way they are
used, machine learning is employed to pick out patterns, tonal
variances and colloquial or non-literal use of language and interpret
what we are trying to express. ML-derived NLP can be seen or
heard in action in virtual assistants such as Apple’s Siri, Microsoft’s
Cortana and Amazon’s Alexa.

Technology
Neural networks
 Algorithmic models structured as hierarchical networks of nodes which all
pass information (data) between themselves, extrapolating more and more
precise meaning and value from it as it passes along the chain.
 Their complex, interconnected nature allows data to be processed far
more comprehensively than traditional, linear algorithms allow, enabling
them more insightful output from big, messy and unstructured datasets.
 The more precise and correct term, artificial neural networks (ANNs), is
often simply shortened to “neural network”, the term for the system of
biological neurons in the animal brain which machine learning attempts to
emulate.
Specialized AI
 The form of AI becoming commonplace in business, scientific research
and our everyday lives – usually in the form of applications designed to
carry out one specific task in an increasingly efficient way.
 This could be anything from giving you tips on improving your fitness by
monitoring exercise patterns, to predicting when machinery will break
down on a production line, to spotting genetic indicators of illness in a
human gene sequence.

Technology
Supervised Learning
 Supervised learning is a term used for machine learning processes
where the output of the algorithm is checked, and the results fed back
to the computer to enable it to know how accurate they are.
 It can then use this knowledge to increase the probability that it will
return with an acceptably accurate result next time around.
 As a simple example imagine an AI fraud detection algorithm designed
to flag up suspicious transactions by a bank.
 In unsupervised learning, data is matched against previous outcomes
to look for patterns in financial transactions, such as their point of
origin, size or time of day they take place, which may indicate they are
suspicious.
 As new suspicious transactions are identified, the algorithm adapts to
“learn” that other features of the newly-identified suspicious
transactions may also be an indicator of fraud.
 In this way, a supervised learning system can learn to identify fraud
from characteristics that were not highlighted in its initial training data
as indicators of fraud.

Technology
Unsupervised Learning
 Unsupervised learning is the flip-side of the coin from supervised
learning and involves giving computers the ability to increasingly
accurately recognize and classify data without needing a human, or
initial training data, to check it if it is right or wrong.
 In unsupervised learning the algorithm only ever sees the input
data, and it classifies it according to patterns that it recognizes from
other input data that it has previously processed.
 This is generally done through a statistical process known as
clustering, where objects (financial transactions to carry on ) are
grouped together according to qualities and attributes that they
share.
 This approach to the problem of data classification has tremendous
potential for developing machines which more closely emulate our
own thought and decision-making processes, but also requires
huge amounts of processing power compared to supervised
learning.

Part – 1
Artificial Intelligence Trailblazers

Using AI to Power the Retail & Business to
Business services of the Future

Alibaba – Using AI to Power Retail & Business
 Alibaba- Chinese company the world’s largest e-commerce marketplace
with US $ 248 billion in transactions ( more than eBay & Amazon combined)
invests in AI and ML , you realize quickly that the list what is doesn’t impact
is much shorter.
 Since 1999 when the company was launched in founder Jack Ma’s
apartment , Alibaba’s core business remains to sell good, but its influence &
operations have expanded to make it one of the Most valuable Tech
companies in the world.
 One of the reasons Alibaba and other Chinese tech companies such
as Tencent and Baidu, collectively known as BAT, are making extraordinary
gains in artificial intelligence is the support, investment and commitment of
the Chinese government to become the dominant AI player in the world.
 With plans to build a US $ 1 Trillion AI Industry by 2030, China is on a path
to overtake the United States as the world’s leader in technology.
 Along with an enormous population that generates critical data to inform AI
algorithms and help make them better as well as a society who is keen on
technological change and not as pushy on regulation, China is fertile
ground to develop AI applications, and Alibaba is moving full throttle ahead.

Along with an enormous population that generates critical data to
inform AI algorithms & help make them better as well as a society
who is keen on technological change & not as pushy on regulation,
China is fertile ground to develop AI applications, and Alibaba is
moving full throttle ahead.
Alibaba invested in 7 research labs that will focus on AI, ML, network
security , natural language processing & more .
Singles’ day leverages AI
 Singles’ Day, once an anti-Valentine’s Day to celebrate being single
has now transformed - thanks to Alibaba and its founder Jack Ma -
into the largest e-commerce event in the world.
 To handle the incredible volume, more than Black Friday and Cyber
Monday combined in one day, Alibaba used automation, robots, AI
and machine learning throughout the process from guiding shoppers
to the products to delivery.
 Here are few AI applications that were used:

 Tmall Smart Selection: This AI-powered algorithm backed by deep
learning and natural language processing helps recommend
products to shoppers and then communicates to the retailers to
increase inventory to keep up with the demand.
 Dian Xiaomi: This AI-powered chatbot can understand more than 90
percent of customer’s queries according to Alibaba and serves more
than 3.5 million users a day. The latest version of the chatbot can
understand a customers’ emotion and can prioritize and alert human
customer service agents to intervene.
 Robots to pack and drones to deliver. More than 200 robots in
automated warehouses can process 1 million shipments each day.
 Once the robots received the orders on Singles’ Day, they packaged
and shipped the goods, and, in some cases, their efficient allowed
same-day shipment. Alibaba also used drones for some deliveries.
These technological advances that helped power Singles’ Day last year resulted in $25
billion in sales, up from $17.8 billion in sales for Single’s Day 2016 and more than the
GDP of Iceland and nearly $20 billion more than Cyber Monday in the U.S.

Turning physical shops into “ Smart Stores”
 In addition to turning some physical shops into “smart stores” for
 Singles’ Day, Alibaba is introducing “digitalization in a box” under its
Tmall brand to smaller retailers.
 Alibaba has redone about 1 million mom-and-pop shops and 100
superstores across China where retailers get all of their goods through
Alibaba's platform and use its Alipay app.
 These efforts allow Alibaba to create self-contained, AI-infused retail
models that combine e-commerce and brick-and-mortar into a mixed
experience.
 Alibaba gives retailers a relatively low cost to join and provides the
operating systems for stores to go digital.
 According to the McKinsey Global Institute, 42% of global ecommerce
transactions took place in China, more than Japan, France, Germany,
the UK and the U.S. combined.
 Many believe that it will only be those retailers who embrace digitization
who will survive, and Alibaba is providing the structure to make it
possible.

Investment in SenseTime, DAMO Academt, & Research
 As China’s largest R&D spender, Alibaba is the largest single
investor in SenseTime; an AI start-up known for its facial-
recognition technology, that launched an AI lab in Hong Kong.
 The lab hopes to "advance the frontier of AI" by supporting
other startups as they commercialize their AI tech and develop
ideas and products. Researchers and other industry
participants can also collaborate with start-ups in the lab.
 Alibaba plans to spend $15 billion over three years on DAMO
(discovery, adventure, momentum, and outlook) Academy.
While AI is the most significant research project for DAMO, it
will also have research groups for emerging technologies such
as blockchain, computer security, quantum computing and
fintech.

City Brain: AI Control For Cities
 With its City Brain project, Alibaba hopes to help cities run
their operations by artificial intelligence.
 Already improving traffic issues, City Brain uses a cloud-based
system where data about a city and everyone in it is stored
and processed through AI algorithms.
 The project’s success in reducing traffic jams by 15% was
achieved by monitoring every vehicle in the city. Already
successful in Hangzhou, City Brain is going to Malaysia next.
 In addition to these examples, Alibaba uses AI to optimize its
supply chain, build products and drive personalized
recommendations.
 Ultimately, Alibaba aspires to be the tech giant to provide
cloud-based AI which would make AI available to anyone with
a computer and internet connection as well as an AI chip
available through the cloud.

 Alibaba is China’s biggest investor in R & D , which
has given it a strong start in the race to become the
world leader in AI.
 Its model for rolling out AT to million of customers &
businesses is to deploy its services through the
cloud. This cuts customer risk & infrastrucutre cost,
while giving Alibaba access to valuable data about
how its customer behaves.
 By applying technology designed to drive sales at
its retail portals to other problems in business and
society, it identifies new use case for AI, within &
outside its established business operations
Key Challenges, Learning points & Takeaways

Maximising the Potential of AI

 Alphabet – US based
Multinational
 Core -Internet services ,
Technology & Life science
 Main business
 Google -Internet search
giant google,
 Verily ( life sciences)
 Waymo ( Self –driving
technology) ,
 Nest ( smart home
company) ,
 Deep mind ( AI).
Alphabet & Google

 Google- widely used search engine is peppered with AI
 Text, Voice or image search – queries processed by smart, self-teaching
systems.
 Text & Voice search uses natural language processing- Every word
entered relates to other word it is used with.
 Google image search uses computer vision to recognize the content of
image data.
 Deep learning algorithms allow it to become increasingly good at
recognizing and labelling different elements of the pictures.
 AI processed the query and decides what the user wants & matches
against its directory online content- web pages, images, videos or
documents.
 Simple google search involves – great deal of complex, fast AI
calculations.
 Google uses AI for Security, Gmail accounts, adwords ( businesses to
pay for ads)
How Alphabet uses AI ?

AI Personal Assistant
 PA using Voice assistants – Google home, Amazon Alexa, Apple Siri
 Limitations- they can respond well to basic, relatively short sentences &
commands.
 They are still infants compare to real human being, since they do not
have enough data.
 This is changing very fast – Google’s Duplex tech is leading this change.
Language Translation
 Machine learning has enables to teach a computer to speak one
language – it can teach itself to speak any language.
 Google’s language translation service uses deep learning to break
languages down to their fundamental building block.
 Google uses deep neutral networks to refine its algorithms.
 Has built features – Pixel Bud headphones leading to real time
translations directly through head phones
Personal Assistants & Language Translation

 No Steering wheels or driver controls
 Designed for new age urban motoring
 Waymo’s service is aimed at the ride-sharing networks for smart cities
Self-Driving Cars

Other segments
Captioning Millions for Videos – Uses machine learning natural
language algorithms for creating subtitles for the hard hearing .
Diagnosing Disease – AI – deep learning technology is extensively
used in the medical field. 3d Infrared scans for diagnosing eye
conditions. Uses two deep learning algorithms – Compares what is
normal and indicative problem and other one based on medical data
.
Google brain – Vast data generated by internet will unlock the
usefulness through machine learning & deep learning.
Deep mind- Neural net simulations of the brain which is trained to
play games.

Alphabet & Google clearly believe that AI is the Launchpad
that will drive the next wave of transformative computer
technology.
As well as this, they believe the social impact of this next wave
will be even greater than that of previous waves- including the
development of the internet.
Having more data than anyone else is a key advantage, which
has enabled Alphabet to continue to develop first-in-class
services from search to ad serving, language translation,
speech processing , smart homes and autonomous driving.

Using Deep Learning to Drive Business
Performance

How does the Amazon Alexa Works ?
 “Alexa, what’s the weather going to be like today.”
 It’s taken decades for scientists to understand natural human speech
to the point where voice-activated interfaces such as Alexa, the
natural language processing system by Amazon, are sufficiently
enabled to be successfully accepted by consumers.
 Alexa is who talks to users of Amazon’s Echo products including the
Echo, Dot and Tap, as well as Amazon Fire TV and other third-party
products.
 Even since 2012, when the patent was filed for what would ultimately
become Amazon’s artificial intelligence system Alexa, there has been
tremendous growth in capabilities and the credit for that growth goes
to machine learning.
 For something that we do every day without giving it any thought,
conversation between machines and humans is complex. So, how did
Amazon and others in the space such as Google, Apple and Microsoft
crack the code?
Amazon

 Over 30 million smart speakers were sold globally last year, and
this number is expected to grow to nearly 60 million this year.
 While Amazon remains the industry leader in smart speakers
selling about 20 million devices last year, others (especially
Google) are also growing and starting to catch up.
 There are nuances to each, but let’s look “under the hood” of an
Echo to see how Alexa works.
 While there is some capability contained in the Echo cylinder
such as speakers, a microphone and a small computer that can
awake the system and blink its lights to let you know it’s
activated, its real capabilities occur once it sends whatever you
have told Alexa to the cloud to be interpreted by Alexa Voice
Services (AVS).
ABC of Alexa

 So, when you ask Alexa, “What’s the weather going to be like today,” the
device records your voice. Then that recording is sent over the Internet to
Amazon’s Alexa Voice Services which parses the recording into commands it
understands. Then, the system sends the relevant output back to your device.
 When you ask about the weather, an audio file is sent back and Alexa tells
you the weather forecast all without you having any idea there was any back
and forth between systems.
 What that of course means is that if you lose internet connection Alexa is no
longer working.
 The skills Echo has out of the box are impressive to most of us, but Amazon
allows and encourages approved developers free access to Alexa Voice
Services so they can create new Alexa skills to augment the system’s skill-set
just as Apple did with the app store.
 As a result of this openness, the list of skills that Alexa (currently over 30,000)
can help with continues to grow rapidly. Users can, of course, purchase
products from Amazon, but they can also order pizza from Domino’s, hail a
ride from Uber or Lyft, control their light fixtures, make a payment through the
Capital One skill, get wine pairings for dinner and so much more.
ABC of Alexa

 Data and machine learning is the foundation of Alexa’s power,
and it’s only getting stronger as its popularity and the amount of
data it gathers increase.
 Every time Alexa makes a mistake in interpreting your request,
that data is used to make the system smarter the next time
around. Machine learning is the reason for the rapid
improvement in the capabilities of voice-activated user interface.
 For example, Google speech was able to improve its error rate
tremendously in a year; now it recognizes 19 out of 20 words it
hears.
 Understanding natural human speech is a gargantuan problem,
and we now have the computing power at our disposal to make
it better the more we use it
Constantly Learning from Human Data

 Amazon- First one to harness the power of predictive
analytics . AI promises more accurate prediction.
 Amazon has built a Corporate strategy – Flywheel to
encourage distribution of energy, momentum and data
generated by AI through out the network of business
operations.
 Alex voice assistant and Amazon Prime Air drone delivery
services through deep learning capabilities with engine
algorithms.
 Amazon leases its machine learning and deep learning
knowledge /technology as a service through AWS platform .

Apple
Integrating AI into Products & Protecting User
Privacy

 World’s largest information technology company by revenue.
 Products – IPhone, IPad, Macs, Apple watch , Apple TC –software and
services.
 Valued at US $ 1 trillion in 2018.
 AI strategy centers around its devices- pioneered of In-device AI
technology leading to superior security and user engaging
experiences.
How does Apple use AI ?
 Vision- Powerful handheld devices capable of running their own
machine learning on datasets gathered via their own array of sensors.
 This is against the vision of other tech companies – future dominated
by cloud computing & low powered terminals.
 Iphone-X Custom designed chip designed for carrying out the neural
net calculations needed for deep learning.
 This leads to faster processing of Face ID logins, features in the
camera ( can add silly effects) , augmented reality and battery life.
Apple

Smarter Apps
 Significant credit goes to App Store.
 App ecosystem keep customers coming back to Apple year after
year because of integrated AI to 3rd party apps.
 E.g. – App Homecourt – assist with refereeing amateur basketball
games. Point the camera at a game & machine learning will tag the
players in the game, logging when they pass & shoot, as well as
recording their position on the court.
 Done through computer vision technology running on the device.
Natural Language processing
 Through Siri – Introduced location signals into the training data,
giving Siri access to localized datasets, including place names and
small businesses.
 Today Siri gives users more accurate data results when they search
for information.
Apple

 AI is very much at the heart of Apple’s strategy, which is to
build it into the fabric of its devices & supporting services.
 Apple is prioritizing user privacy over an ability to pump all
data into the cloud to train algorithms on bigger data sets.
 It is also promoting the use of its proprietary machine
learning platform Create ML to make apps that will only work
on its devices, creating exclusivity within its own app
ecosphere.

Machine Learning for Search Engines &
Autonomous Cars

 At the beginning of 2017, Chinese tech company BAIDU, the largest provider
of Chinese language internet search as well as other digital products and
services, committed to emerging business sectors such as AI & Machine
Learning .
 Since China has over 800 million internet users, almost twice the U.S.
population, Baidu’s data set is capable of fueling AI algorithms to make them
even better.
 With this focus on artificial intelligence, Baidu is exploring some very
intriguing applications for artificial intelligence and machine learning including
in their offices where facial recognition technology makes standard ID cards
unnecessary and allows you to order tea from a vending machine.
 They recruited top AI talent including one of the world’s most notable AI
pioneers Li Qi, who was previously a Microsoft executive before he became
Baidu’s COO in January 2017. & he stepped down in July 2018 for personal
reasons.
 Although he was only at Baidu for a short time, he helped chart a clear
strategy for the company’s AI operations that will continue. Here are a few
ways Baidu uses artificial intelligence and machine learning.
BAIDU

 Baidu can leverage its expansive data set, its voice assistant called
DuerOS has accumulated more conversation-based skill sets than
Alexa, Siri or Cortana.
 Partnering with other tech companies is one way Baidu hopes to
accelerate innovation.
 They have teamed up with more than 130 DuerOS partners, and the
voice assistant is in more than 100 brands of appliances such as
refrigerators, TVs, and speakers.
 Since homes in India, Japan, Europe, and Brazil are more like homes
in China, there may be better opportunities for DuerOS to globalize
since Alexa, Cortana and Echo are optimized for American
households.
 At CES 2018, Baidu debuted its DuerOS-Powered smart screen called
Little Fish VS1.
 This technology can recognize and respond to individual faces.
DuerOS is Baidu’s voice assistant

 Even though automated driving is currently against the law in
China, Baidu is working on autonomous-driving technology.
 Through the program called Apollo, Baidu’s Artificial intelligence
technologies are made available to car makers for free as a
brain for their cars. In exchange, Baidu gets access to the data
to make their algorithms smarter.
 It is hoped that Apollo will give any car manufacturer a fair shot
at creating a viable product, just like Android did for smartphone
makers.
 Even with this accelerated approach, it is expected that fully
autonomous cars won’t be in production until 2020-2021.
Self Driving Cars

 Unlike its competitors, Baidu was steadfast in its commitment to
desktops and missed the shift to mobile.
 To survive, Baidu needed a new strategy and artificial
intelligence technology provided just the platform to turn the
business around.
 That’s one of the reasons Baidu has committed so aggressively
to AI investment.
 Today, AI products and services are priorities to make them the
core of the company’s future.
 Now, they are partnering with Huawei to develop an open
mobile AI platform to support the development of AI-powered
smartphones and Qualcomm to optimize its DuerOS for IoT
devices and smartphones using Qualcomm’s Snapdragon
Mobile Platform
Mobile partners to accelerate AI-powered devices

 Has developed a handheld device capable of generating deep
learning translation between English, Mandarin, Chinese &
Japanese.
 Aimed at tourist market- assisting users to navigate their way
around foreign cities – ordering food, using public transport etc.
 Uses deep learning natural language processing algorithms.
 It is on cloud.
Real Time Translation

 The huge population base with more 50% online has helped
the company to collect vast data of consumer profile &
behaviours. This is used to streamline services, as well as
sell to advertisers to allow them to more accurately target
their campaigns.
 Baidu offers AI services to businesses to enable them to
develop & release their own AI-powered applications under
its Baidu Brain framework.
 Making strategic partnership with China’s largest smartphone
manufacturer- Huawei to incorporate AI inside smart phones.
 Baidu has China’s and possibly the world’s most advanced
autonomous vehicle program with cars powered by its Apollo
technology expected to bring a level 4 autonomy to the roads
soon

Using AI to improve Social Media Services

 Facebook builds its business by learning about its users and packaging
their data for advertisers.
 It then reinvests this money into offering us new, useful functionality –
currently video and shopping - which it also uses to learn even more
about us.
 As the way it enables communication & conversation between people has
proven to be hugely valuable to us, it has become a magnet for a huge
amount of data about us – who we are, where we spend our time and
what we like.
 The challenges for Facebook’s data scientists who have to try to make
sense of this is that much of this data is very messily unstructured.
 2.2 billion people use the FB Social media platform. No. of comments –
510,000 & 293,000 status update / minute
 With 1.2 billion people uploading 136,000 photos and updating their status
293,000 times per minute, until recently Facebook could only hope to
draw value from a tiny fraction of its unstructured data – information which
isn’t easily quantified and put into rows and tables for computer analysis.
Facebook- Using AI

 Deep Learning is helping to play a part in changing that. Deep
Learning techniques enables machines to learn to classify data by
themselves.
 A simple example is a deep learning image analysis tool which
would learn to recognize images which contain cats, without
specifically being told what a cat looks like.
 By analyzing a large number of images, it can learn from the
context of the image – what else is likely to be present in an image
of a cat?
 What text or metadata might suggest that an image contains a
cat?.
 That’s the basic principle of why Deep Learning (DL) is useful to
Facebook, and as DL algorithms become more sophisticated they
can increasingly be applied to more data that we share, from text to
pictures to videos
Facebook- Using AI- Deep Learning

 Facebook uses a DL application called DeepFace to teach it
to recognize people in photos.
 It says that its most advanced image recognition tool is more
successful than humans in recognizing whether two different
images are of the same person or not – with DeepFace
scoring a 97% success rate compared to humans with 96%.
 It was fed more than 4 million facial images to train it how to
recognize individual facial elements.
 Users can keep track of where photos of themselves are
cropping up on the site.
 FB confirms that its facial recognition algorithms have a
success rate of 97.35% - very close to human-level accuracy.
Facial Recognition

 Facebook uses a tool it developed itself called DeepText to extract
meaning from words we post by learning to analyze them contextually.
 Neural networks analyze the relationship between words to
understand how their meaning changes depending on other words
around them.
 Because this is semi-unsupervised learning, the algorithms do not
necessarily have reference data – for example a dictionary –
explaining the meaning of every word. Instead, it learns for itself based
on how words are used.
 This means that it won’t be tripped up by variations in spelling, slang
or idiosyncrasies of language use. In fact, Facebook say the
technology is “language agnostic” – due to the way it assigns labels to
words, it can easily switch between working across different human
languages and apply what it has learned from one to another.
 At present the tool is used to direct people towards products they may
want to purchase based on conversations they are having.
Understanding Text

 The vast amount of information we share about our lives on
FB means that the company has access to more of our
personal data than just about anyone else.
 FB has leveraged this to build features that keep us coming
back to the site to share more data as well as match us with
advertisers whose products we might want to buy.
 All these data- including our photos & text- has been
invaluable to FB when it comes to training its facial
recognition & natural language processing algorithms.
 Unprecedented levels of insight into our lives means it can
make increasingly accurate predictions about the users- from
what users want to buy to whether we are thinking about
killing ourselves.

Cognitive Computing Helps Machines Debate with
Humans

 The IBM algorithm Deep Blue beat chess champion Garry Kasparov in
1997.
 It was 2011 when IBM’s Watson won the game show Jeopardy.
 Shortly after, the IBM Research team was ready to go beyond game
playing and began to brainstorm the next feat to challenge an artificial
intelligence algorithm.
 They decided to create an AI algorithm that would be trained on the art
of debate.
 Recently a small group of viewers got to see the IBM Project Debater’s
public debut and its first two debates, when it went head-to-head with
Israeli debaters Dan Zafrir and Noa Ovadia on increased investment in
telemedicine and government subsidies for space exploration
respectively.
 From all accounts, IBM Project Debater was a formidable opponent
and surprised many with its ability to make human-like arguments. It
even swayed more audience members to its position on telemedicine
that Zafrir did.
How Does IBM use AI

 This project was the latest in IBM Research’s goal to build a system “that
helps people make evidence-based decisions when answers aren’t black-
and-white.”
 Debate not only helps us convince others of our opinion, but it can help us
understand and learn from other’s views.
 By training machines in this way, it is hoped that in the future, AI algorithms
will be able to help humans make important decisions regularly.
 IBM Project Debater doesn’t just search its database of millions of articles
from well-known newspapers and magazines—its corpus—but it has AI
technology that can “work with humans to discover, reason and present
new points of view.”
 The IBM Research team was able to create an algorithm with the ability to:
 Generate an opinion driven by data
 Listen and understand an opponent, parsing out the critical bits of data
from flowing narrative
 Express the situation and arguments with concise language and
complete human-like sentences
How Does IBM use AI

 One of the impressive abilities IBM Project Debater exhibited was the
combination of AI techniques it relied upon to solve many problems
and join them together in a solution.
 Now that IBM Research succeeded in this first debate, the team needs
to determine practical applications of this technology that they can sell.
 “Project Debater’s underlying technologies will also be commercialized
in IBM Cloud and IBM Watson in the future.”
 Now that AI has gone beyond playing games to learning the art of
persuasion and debate; it has proven that it can handle the "gray area”
and nuances of human interaction and not just follow clear-cut rules.
 “From IBM perspective, the debate format is the means and not the
end. It's a way to push the technology forward and part of our bigger
strategy of mastering language,”
 It was an impressive debut, and it will be intriguing to see what’s up
next.
Practical application of the Technology

 Thousands of businesses are using IBM Watson to take
advantage of AI- Customer relations, Chatbots & Medicine.
 IBM strategy is to breakdown communication barriers
between people & machines to harness their potential.
 IBM uses gameplay-that its cognitive systems are capable of
learning to solve puzzles in the same way that humans do.
 Project Debater represents AI evolving past its current ability
to answer questions & towards being able to engage in
natural human conversations.

 Often referred to as the Amazon of China, JD.com started in 1998
as a brick-and-mortar store in Beijing, but it has aspirations to be
the world’s leading e-commerce retailer.
 Based on its tremendous growth, it might not take long for the
company to get there. Richard Liu, the company’s founder, CEO &
Chairman has even gone so far to predict his company won’t need
humans and said “I hope my company would be 100% automation
someday…no human beings anymore, 100% operated by AI and
robots.”
 JD.com and its competitors such as Amazon, Alphabet, Tencent,
Alibaba and more are not only racing to be the world's largest e-
commerce business but to create the operating system for retail in
the future. JD.com is driving business with AI, big data, and
robotics while building the retail infrastructure for the 4th industrial
revolution
JD.Com - AI

 To handle delivery, logistics and supply chain across vast retail
network.
 Shanghai fulfilment center- 200000 orders per day – employs 4
people.
 Robots- Powered with machine learning, move crates of products to
conveyor belts, packed by other set of robots and gets despatched.
 This has enabled next day delivery services anywhere in China for 1.3
Billion people- 10 million Km of territory.
 Working on same day delivery.
 Has a chatbot that is capable of producing automated poetry when
items are purchased as gifts.
 Partnered with China’s social media giants- Tencent and Baidu to
integrate messaging and image sharing apps.
 AI is used to match users, based on profile data with items sold by
JD.com
What does JD.Com use AI for ?

 JD.com is not only investing in the technology of tomorrow, but they are also
applying today's technology into its operations in many ways from smart
warehouses to drone delivery.
 Here are just a few ways JD.com uses AI, big data, and robotics in operations
today:
 Automated warehouses: While JD.com’s warehouses aren’t entirely
autonomous, they are taking action currently to automate everything they
possibly can.
 Robots: Some of the company’s most advanced robots work in its 500
warehouses. They stack products on shelves and pack and ship
merchandise to send out to consumers.
 Drones: JD.com has used Drones to deliver products across China since
March 2016- close to proximity of drone stations- furthest being 15 Km.
 They use drones of various shapes and sizes, and they are currently working
to build a drone that can carry up to five tons.
 Autonomous trucks deployed by the company have accumulated 17,000
hours of road driving experience- Human driver is a must in cities.
 Working towards unmanned trucks
Automated deliveries By Air & Road

 The company is testing out facial recognition software at its
headquarters that would allow shoppers to take their merchandise
out of a store without stopping to pay; payment is controlled
through facial recognition.
 Customers can start signing from their smartphones and using the
cameras to upload images of their faces ( HD) for identification. (
Machine learning)
Facial Recognition Technology
Smart Fridges
Has announced Smart Fridges that uses
camera equipped with image recognition
technology.
Cameras can scan items in the fridge and
inform the expiry dates.
It will inform to the smart phone items
which are running low for ordering.

 JD.com founder has said that he hopes to see his company’s
human staff reduced from 160,000 to 80,000 within next 10 years.
While he says many will be retrained, it seems retaining human
jobs comes secondary to driving efficiencies & improving
customer experience.
 Driving efficiency within its operations and supply chain is
JD.com’s primary motivation for rolling out AI. Automated
warehouses , delivery networks & retail outlets all form a part of
this plan.
 JD.com is also partnered with social media providers to allow
them to use data on their customers for AI-driven precision
marketing campaigns carried out entirely through their social
apps.
 Starting out as a brick-and-mortar retailer, JD.com is blurring the
boundaries between online & offline shopping through a drive to
introduce e-commerce technology into its physical stores.
Key Challenges, Learning Points & Takeaways

Making AI Part of the Fabric of everyday Life

 AI tools contained in Office 365.
 PowerPoint is capable of giving design tips based on how it
observes the user working.
 Work uses AI to suggest meanings, alternate phrases & check
spelling, grammar & punctuation.
 Azure cognitive services offers “ pre-built” machine learning
solutions for speech recognition, text analysis, computer vision
& language translation.
 Another tool that has the potential to be very useful is
Sketch2Code- capable of generating working HTML websites
from simple sketches.
 Microsoft offers online AI school. This is the collection of
resources that covers the basics of what AI can do and how to
start using it.
How does Microsoft use AI

Underwater Data Centers
 Cloud based AI requires a lot of network bandwidth.
 Project Natick, which involves submerging data centers under the ocean to
closed to coastal cities.
 It is the size of the shipping containers & self contained.

 IBM vision is that AI will eventually become simple a part of
fabric of everyday life- much like computers and internet.
 To achieve this company is building tools & services that let
other businesses carry out machine learning through their
Azure cloud infrastructure.
 AI functionality in its mainstream office productivity software
used by millions – making jobs quicker & easier – Machine
learning.
 Microsoft has partnered with businesses of all shapes and
sizes to roll out AI solutions & is now diving into
reinforcement learning with its acquisition Bonsai.

Using AI To power Wechat & Health care

 Tencent is a Chinese tech company founded in 1998 and based
in Shenzhen that hosts 55% of China’s Mobile Internet usage on
its platforms.
 Its mission is to “become the most respected internet
enterprise.”
 The company is China’s biggest social network company with 1
billion users on its app WeChat and 632 million monthly user
accounts on social networking platform Qzone, is worth more
than Facebook and has extended beyond instant messaging (its
product is QQ) and social networking to gaming, digital
assistants, mobile payments, cloud storage, education, live
streaming, sports, movies and artificial intelligence.
 The company’s dedication to artificial intelligence is evident in
one of its slogans, “AI in all.”
Tencent

 Stands out for the advancements it has made in facial
recognition technology.
 3 Chinese provinces- citizens are allowed to verify their identity
through WeChat, Digital ID cards,.
 Technology used in video games
 It has trained software robots to become so good at strategy
game Starcraft 2 that it can beat the computer team’s AI bots on
the highest difficulty setting.
How does Tencent use AI
 Tencent is a top investor ( reported to be $120 Million) in robotics start-up
UBTech, a firm that focuses on humanoid robots.
 Quite possibly UBTech’s most famous contribution is Walker, a bipedal robot
unveiled at the 2018 Consumer Electronics Show that can walk downstairs.
 Of the Chinese tech firms known collectively as BAT (Baidu, Alibaba and
Tencent), Tencent participated in the greatest number of AI equity deals and
made the most AI investments in the United States.

 Healthcare AI is a main priority for Tencent, based on the
company’s investments and AI partnerships.
 China would like to be a world leader in personalized medicine
using AI. More than 38,000 medical institutions have a WeChat
account and 60% of those institutions allow patients to book
appointments online.
 Additionally, there are 2,000 hospitals that accept WeChat
payment.
 These services allow Tencent to collect valuable consumer data
that helps train AI algorithms.
 In a recent partnership with Babylon Health, WeChat users will
have access to a virtual healthcare assistant.
 Pushing the envelope even further, Tencent invested in iCarbonX,
a company that aims to develop a digital representation of
individuals to help perfect personalized medicine.
Medical Technology

 Tencent is one of the China’s largest investors in AI & looks
for opportunities to capitalize on AI across all of the industries
in which it operates.
 Leaders in Natural language processing, image recognition &
machine learning.
 Technology has huge implications for gaming and also create
new gameplay challenges for players.
 Successful in building AI into healthcare systems, helping
surgeries and hospitals to run smoothly, and assisting
doctors with diagnosing and treating illness.

Mail your comments to
ramaddster@gmail.com
End of Part -1
Will continue the summary in Part - 2

Artificial intelligence in practice- part-1

Recommended

Recommended

More Related Content

What's hot

What's hot (20)

Similar to Artificial intelligence in practice- part-1

Similar to Artificial intelligence in practice- part-1 (20)

More from GMR Group

More from GMR Group (20)

Recently uploaded

Recently uploaded (20)

Artificial intelligence in practice- part-1