CS249: Artificial Intelligence

Asif Ekbal
CSE Dept.
IIT Patna
Lecture–1: Introduction

1/17/2023
 THE MOST EXCITING DISCIPLINE in

Today’s World

1/17/2023
Artificial Intelligence: Not merely a humanoid
robot

AI, as often portrayed, in science fiction is NOT only robots or other

humanoid beings
○ who are friendly and serve humans or,
○ turn evil and want to kill all humans to take control of our
planet

Fei-Fei Li, Director of Stanford AI lab, claims that the myth of the
terminator coming next door is, in fact, a real crisis for the development of
the AI field as it highlights the public misreading of the technology but also
reveals the fear of what are the intentions of the people behind the
technology (2018)

3
● AI is, in fact, an ever-evolving term which is one of the
reasons that it means very different things to different people

● Artificial Intelligence is hard to define because the field has

been redefined continuously with the advances of technology
and the ambiguity of what we consider as “intelligent”

A better understanding of AI is crucial to its future

development and progress

4
 Why study AI?

Search engines
Science

Medicine/
Diagnosis

Labor
Appliances What else?

1/17/2023
Honda Humanoid Robot

Walk

Turn

http://world.honda.com/ASIMO/
Stairs
1/17/2023
 Sony AIBO

1/17/2023
Natural Language Question Answering

http://aimovie.warnerbros.com http://www.ai.mit.edu/projects/infolab/
1/17/2023
 Example: AI at NASA

1/17/2023
1/17/2023
 Examples: AI at Google

1/17/2023
Finding Canonical Images

1/17/2023
Compare low-level features

1/17/2023
 Induced Graph

1/17/2023
Route Finding

1/17/2023
COVID-19 Trend

1/17/2023
1/17/2023
1/17/2023
 Google Sets
tigers, lions, bears, cats, dogs, birds, horses, reptiles,
fish, pet breeders, rabbits, pet health, pet food, pet
supplies, exotic pets, small mammals, iguanas,
animal medications, pet rescue, pet adoption,
veterinary jobs, pets general, elephants, livestock,
insects, other, small pets, marine life, spiders,
amphibians, aquariums, pets, wolves, monkeys,
flowers, pigs, wildlife, dolphins, frogs, giraffes,
turtles, butterflies, sheep, ferrets, dinosaurs, farm
animals

1/17/2023
Statistical Machine Translation

SEHR GEEHRTER GAST! DEAR GUESTS,

KUNST, KULTUR UND ART, CULTURE AND
KOMFORT IM HERZEN LUXURY IN THE HEART
BERLIN. OF BERLIN.

DIE ÖRTLICHE THE LOCAL VOLTAGE

NETZSPANNUNG BETRÄGT IS 220/240 VOLTS 50 HZ.
220/240 VOLT BEI 50
EN
HERTZ.
DE
1/17/2023
Statistical Machine Translation

n Align

KUNST, KULTUR UND KOMFORT IM HERZEN BERLINS.

ART, CULTURE AND LUXURY IN THE HEART OF BERLIN.

1/17/2023
Statistical Machine Translation

n Align

KUNST, KULTUR UND KOMFORT IM HERZEN BERLINS.

ART, CULTURE AND LUXURY IN THE HEART OF BERLIN.

1/17/2023
Statistical Machine Translation

n Align
KUNST, KULTUR UND KOMFORT IM HERZEN BERLINS.

ART, CULTURE AND LUXURY IN THE HEART OF BERLIN.

1/17/2023
 Statistical Machine Translation
n Align
KUNST, KULTUR UND KOMFORT IM HERZEN BERLINS.

ART, CULTURE AND LUXURY IN THE HEART OF BERLIN.

1/17/2023
 Statistical Machine Translation

KUNST, KULTUR UND KOMFORT IM HERZEN BERLINS.

ART, CULTURE AND LUXURY IN THE HEART OF BERLIN.

1/17/2023
English-Hindi and Challenges
n The grandfather kicked the bucket after suffering from cancer.
n This job is a piece of cake
n Put the sweater on
n He is the dark horse of the match

Google Translations of above sentences:

दा दा कै ं सर से प ी ड ़ ि त ह ो न े के ब
ा द बा ल् ट ी ला त म ा र ी .
इ स का म के के क का ए क ट ु कड ़ ा ह ै .
स् व े ट र प र र खो .
व ह मै च के अं ध े र े घो ड़ ा है .

26
English-Hindi and Challenges
n Bengali: চঞ্ সরকার বাড়িতে আতে
English: Government is restless at home. (*)
Chanchal Sarkar is at home

Amsterdam airport: “Baby Changing Room”

n Hindi: द ै न ि क द ब ं ग द ु न ि य ा
English: Daily domineering world
Actually name of a Hindi newspaper in Indore

n High degree of overlap between NEs and MWEs

n Treat differently - transliterate do not translate
27
What Else is AI?

1/17/2023
1/17/2023
1/17/2023
1/17/2023
1/17/2023
1/17/2023
AI as the forcing function
n Time sharing system in OS
n Machine giving the illusion of attending
simultaneously with several people

n Compilers
n Raising the level of the machine for better
man-machine interface
n Arose from Natural Language Processing (NLP)
n NLP in turn called the forcing function for AI

1/17/2023
 AI is Real Fun!

1/17/2023
What is AI?

1/17/2023
From Wikipedia
Artificial intelligence (AI) is the intelligence of machines and the branch of computer
science that aims to create it. Textbooks define the field as "the study and design of
intelligent agents"[1] where an intelligent agent is a system that perceives its environment
and takes actions that maximize its chances of success.[2] John McCarthy, who coined the
term in 1956, [ 3 ] defines it as "the science and engineering of making intelligent
machines."[4]

The field was founded on the claim that a central property of humans, intelligence—the
sapience of Homo sapiens—can be so precisely described that it can be simulated by a
machine. [5] This raises philosophical issues about the nature of the mind and limits of
scientific hubris, issues which have been addressed by myth, fiction and philosophy since
antiquity.[6] Artificial intelligence has been the subject of optimism,[7] but has also suffered
setbacks[8] and, today, has become an essential part of the technology industry, providing
the heavy lifting for many of the most difficult problems in computer science.[9]

AI research is highly technical and specialized, deeply divided into subfields that often fail to
communicate with each other.[10] Subfields have grown up around particular institutions,
the work of individual researchers, the solution of specific problems, longstanding
differences of opinion about how AI should be done and the application of widely differing
tools. The central problems of AI include such traits as reasoning, knowledge, planning,
learning, communication, perception and the ability to move and manipulate objects.[11]
General intelligence (or "strong AI") is still a long-term goal of (some) research.[12]
1/17/2023
Acting Humanly: The Full Turing Test
n Alan Turing's 1950 article Computing Machinery and Intelligence discussed
conditions for considering a machine to be intelligent
n “Can machines think?”  “Can machines behave intelligently?”
n The Turing test (The Imitation Game): Operational definition of intelligence

• Computer needs to posses: Natural language processing, Knowledge

representation, Automated reasoning, and Machine learning
• Problem: 1) Turing test is not reproducible, constructive, and amenable to
mathematic analysis. 2) What about physical interaction with interrogator and
environment?
• Total Turing Test: Requires physical interaction and needs perception and
actuation.

1/17/2023
What would a computer need to pass the Turing test?
n Natural language processing: to communicate with examiner
n Knowledge representation: to store and retrieve information provided before or
during interrogation
n Automated reasoning: to use the stored information to answer questions and to
draw new conclusions

n Machine learning: to adapt to new circumstances and to detect and extrapolate

patterns

n Vision : to recognize the examiner’s actions and various objects presented by the
examiner
n Motor control: to act upon objects as requested
n Other senses: such as audition, smell, touch, etc.

1/17/2023
Thinking Humanly: Cognitive Science

n 1960 “Cognitive Revolution”: information-processing psychology replaced

behaviorism

n Cognitive science brings together theories and experimental evidence to model

internal activities of the brain
n What level of abstraction? “Knowledge” or “Circuits”?
n How to validate models?
n Predicting and testing behavior of human subjects (top-down)

n Direct identification from neurological data (bottom-up)

n Building computer/machine simulated models and reproduce results

(simulation)

1/17/2023
 Thinking Rationally: Laws of Thought
n Aristotle (~ 450 B.C.) attempted to codify “right thinking”
What are correct arguments/thought processes?
E.g., “Socrates is a man, all men are mortal; therefore Socrates is
mortal”

n Several Greek schools developed various forms of logic:

notation plus rules of derivation for thoughts

n Problems:
1) Uncertainty: Not all facts are certain (e.g., the flight might be
delayed)
2) Resource limitations: There is a difference between solving a
problem in principle and solving it in practice under various resource
limitations such as time, computation, accuracy etc. (e.g.,
purchasing a car)

1/17/2023
Acting Rationally: The Rational Agent
n Rational behavior: Doing the right thing!

n The right thing: That which is expected to maximize the expected

return

n Provides the most general view of AI because it includes:

n Correct inference (“Laws of thought”)
n Uncertainty handling
n Resource limitation considerations (e.g., reflex vs. deliberation)
n Cognitive skills (NLP, AR, knowledge representation, ML, etc.)

n Advantages:
1) More general
2) Its goal of rationality is well defined

1/17/2023
How to achieve AI?
n How is AI research done?

n AI research has both theoretical and experimental sides. The

experimental side has both basic and applied aspects

n There are two main lines of research:

n One is biological, based on the idea that since humans are
intelligent, AI should study humans and imitate their psychology
or physiology
n The other is phenomenal, based on studying and formalizing
common sense facts about the world and the problems that the
world presents to the achievement of goals

n The two approaches interact to some extent, and both should

eventually succeed. It is a race, but both racers seem to be walking.
[John McCarthy]

1/17/2023
Disciplines which form the core of AI- inner circle
Fields which draw from these disciplines- outer circle

Robotics

NLP

Search,
Expert Reasoning,
Systems Machine Learning
Planning

Computer
Vision
1/17/2023
Allied Disciplines
Philosophy Knowledge Rep., Logic, Foundation of AI
(is AI possible?)
Maths Search, Analysis of search algos, logic

Economics Expert Systems, Decision Theory,

Principles of Rational Behavior
Psychology Behavioristic insights into AI programs

Brain Science Learning, cognitive science, Neural Nets

Physics Learning, Information Theory & AI,

Entropy, Robotics
Computer Sc. & Engg. Systems for AI

1/17/2023
AI History
1943 McCulloch & Pitts: Boolean circuit model of brain
1950 Turing’s “Computing Machinery and Intelligence”
1952–69 Look, Ma, no hands!
1950s Early AI programs, including Samuel’s checkers program, Newell &
Simon’s Logic Theorist, Gelernter’s Geometry Engine
1956 Dartmouth meeting: “Artificial Intelligence” adopted
1965 Robinson’s complete algorithm for logical reasoning
1966–74 AI discovers computational complexity Neural network
research almost disappears
1969–79 Early development of knowledge-based systems
1980–88 Expert systems industry booms
1988–93 Expert systems industry busts: “AI Winter”
1985–95 Neural networks return to popularity
1988 – Resurgence of probability; general increase in technical depth
“Nouvelle AI”: ALife, GAs, soft computing
1995 – Agents, agents, everywhere . . . Machine learning comes to age,
web intelligence, smart machines
2003 – Human-level AI back on the agenda
AI State-of-the-art
n Have the following been achieved by AI?
n World-class chess playing

n Playing table tennis

n Solving mathematical problems

n Discover and prove mathematical theories

n Engage in a meaningful conversation

n Understand spoken language

n Observe and understand human emotions

n Express emotions

n …

1/17/2023
AI, NLP and ML: From Past to Present
● AI/NLP based systems have enabled wide-range of
applications
– Google’s powerful search engines
– Alexa, Google’s MT
– Several systems related to healthcare, mental health, processing
social media texts etc.
– Fake news detector and so on
– Chess playing
● Shallow ML algorithms (corresponds to Statistical NLP)
– Used extensively (HMM, MaxEnt, CRF, SVM, Logistic Regression
etc.)
– Requires handcrafting of features
– Time-consuming
– Curse of dimensionality (because of joint modeling of language
models) 48
AI, NLP and ML: From Past to Present
● Deep Learning algorithms
– No feature engineering
– Success of distributed representations (Neural language models)
● Some recent developments
– The rise of distributed representations (e.g., Word2vec, GLOVE,
ELMO, BERT etc)
– Convolutional, recurrent, recursive neural networks,
Transformer, Reinforcement learning
– Unsupervised sentence representation learning
– Combining deep learning models with memory-augmenting
strategies
● Explainable AI

49
AI, NLP and ML: Future

● Artificial General Intelligence :

able to exhibit human intelligence

● Artificial Super Intelligence :

surpasses human intelligence in many aspects

from creativity to general wisdom to problem-solving —

will require machines to experience consciousness

50
News: March 27, 2019
Yoshua Bengio, Geoffrey Hinton, and Yann LeCun received
the

Turing Award-2018 (equivalent to Nobel Prize of

Computing)
n for Modern AI (specifically for deep learning research)

Bengio- University of Toronto and Google

Hinton- University of Montreal
LeCun- Facebook’s chief AI scientist and a professor at NYU
Topics to be covered (1/2)
n Search
n General Graph Search, Uninformed search, informed search

and local search

n Logic
n Propositional Calculus and Predicate Calculus

n Machine Learning
n Basic Ideas, Decision Trees, Bayes Netw o r ks, T e xt

Classification, Neural Networks

n Association Rule Mining

1/17/2023
Topics to be covered (2/2)

n Evolutionary Computation
n Genetic Algorithm, Multiobjective Optimization, Differential

Evolution, Particle Swarm Optimization

n Probabilistic Methods
n Hidden Markov Model

n Introduction to NLP

1/17/2023
Goal of Teaching the course

n Concept building: firm grip on foundations, clear ideas

n Coverage: grasp of good amount of material, advances

n Inspiration: get the spirit of AI, motivation to take up

further work

1/17/2023
Resources
n Main Text:
n Artificial Intelligence: A Modern Approach by Russell &
Norvig, Pearson, 2003.
n Other Main References:
n Principles of AI - Nilsson
n AI - Rich & Knight
n Knowledge Based Systems – Mark Stefik
n Journals
n AI, IEEE Expert, TPMII, IEE TKDE etc..
n Area Specific Journals e.g, Computational Linguistics etc.
n Conferences
n IJCAI, AAAI, ACL etc.

Positively attend lectures!

1/17/2023
Evaluation

n Assignments/projects+ Attendance : 50

n Midterm Exam: 20

n Final Exam: 30

1/17/2023
Instructor

n Dr. Asif Ekbal

Contact:
Email: [email protected]/[email protected]

1/17/2023
Teaching Assistants

n Theory
n Soumitra Ghosh (3rd Year PhD Student, CSE): [email protected]
n Deeksha Vershney (3rd Year PhD student, CSE):
[email protected]
n Mamta Kumari (2nd Year PhD Student, CSE): [email protected]
n Prashant Kapil (4th Year PhD Student, CSE): [email protected]

n Laboratory
n Zishan Ahmad (3rd Year PhD Student, CSE): [email protected]
n Kshitij Misra (2nd Year PhD Student, CSE): [email protected]
n Aizan Zafar (2nd Year PhD Student, CSE): [email protected]
n Kanchan Jha (3rd Year PhD Student, CSE): [email protected]
n Gitanjali Singh (2nd Year PhD Student, CSE): [email protected]

1/17/2023
 Thank you for your attention!

1/17/2023