The document introduces artificial intelligence, machine learning, and deep learning. It discusses supervised, unsupervised, and reinforced learning techniques. Examples of applications discussed include image recognition, natural language processing, and virtual assistants. The document also notes that some AI systems have developed their own internal languages when interacting without human supervision.
3. Artificial Intelligence is the ability of computer systems to
perform tasks that normally require human intelligence, such
as visual perception, speech recognition & decision-making.
A field of study that seeks to explain and emulate intelligent
behaviour in terms of computational processes
Artificial Intelligence is applied when a machine mimics
"cognitive" functions that humans associate with other human
minds, such as "learning" and "problem solving”
Artificial Intelligence
4. Learning denotes changes in a system that
enable a system to do the same task more
efficiently the next time
Machine learning is programming computers
to optimize a performance criterion using
example data or past experience
Deep Learning is a subfield of machine learning concerned
with algorithms inspired by the structure and function of the
brain called artificial neural networks.
Machine Learning
5. Supervised
Learning
• To map a logic
when input and
output is given to
a computer
Unsupervised
Learning
• No labels are
given to the
learning
algorithm,
leaving it on its
own to find
structure in its
input
Reinforcement
Learning
• A computer
program
interacts with a
dynamic
environment in
which it must
perform a certain
goal
Machine Learning Classification
6. Neural Network Architecture. A Brain modelled.
In between the input units and output units are one or more layers of hidden units, which, together, form the
majority of the artificial brain. Most neural networks are fully connected, which means each hidden unit and
each output unit is connected to every unit in the layers either side. The network allows self development hidden
layers and assign weights to these layers from inputs to match with output layers. This is called supervised
learning
7. Architecture in Reinforced learning
During
unsupervised
and reinforced
learning, the
outcome being
a positive or
negative
indicator,
reinforces the
behavior the
network to
promote or
demote a
node
8. Can machine
brains dream?
In a famous experiment by
Google, allowing neural nets
trained to identify images to
run a continuous feedback
loop by linking output to input,
thereby creating a dream like
state for the neural net,
resulted in these remarkable
images.
9. Bots developing their own languages
● Google translate neural net developed its own artificial language to translate context of complete
sentences
● Facebook neural bots assigned with task of learning negotiation eventually learned to lie by themselves
and also developed their own language. Facebook eventually shut down the program.
● In both the situations, humans were incapable of understanding these artificial languages created by bots
for their own communication.
11. AI as a service
Application Business Context
Vision Image processing algorithm to identify , caption and
moderate pictures
Knowledge Map complex information and data to solve tasks such as
intelligent recommendation and semantic search
Language Allow apps to process natural language with pre-built scripts
and learn how to recognize what users want
Speech Convert spoken audio into text
Search Search APIs to your apps and harness ability to comb
billions of webpages, images, videos
Examples:
Microsoft Cortana
IBM Watson used in domains:
• Retail
• Financial Services
• Education Sector
• Health Sector
Every Google application:
• Google Search
• YouTube
• HDR+
• Google Drive
14. Natural Language Processing
● Ability of machines to
understand and interpret
human language the way it
is written or spoken
● Applications in solving
business problems by using
NLP in Big Data
● Application in Log Analysis
and Log Mining to extract
useful information and
knowledge
15. Communication
Applications
Skype Translator for
real time language
interpretation
Automatic Speech
Recognition & Text To
Speech applications in
Search Engines
Customer Review
Improve customer
satisfaction by
analyzing large volume
of customer reviews
Suggest and target
more relevant
products by Big Data
analysis through NLP
Virtual digital
assistants
Online Purchases,
Music Streaming,
Providing Surroundings
Information
Apple’ Siri, Google
Assistant, Amazon
Alexa, Microsoft’s
Cortana
Business Applications NLP
16. Image Recognition Use Cases
• Emerging field of AI that analyses images and retrieves
information about them real time
• Camera based Google translate
• Face detection and object identification for sorting photo
libraries
• Google Photos
• Assistance for Online Shopping
• Point & Shop in Amazon Flow