Artificial Intelligence (Part A)

Intelligence is the most precious gift which most of the humans have, and most of them make good use of it, though may not always be at the right time. Humans are so intelligent that we go forth with teaching this skill to other people and even our beloved machines. In this course I try to make the students intelligent, so that they may make the lame looking machines intelligent and do some awesome stuff. Thinking about thinking is a weird thing to do, which however lays the foundation of making machines intelligent by incorporating the same thinking fundamentals. And attributed to their enormous storage capacity and speed to process information, both much higher than the practically what the humans exhibit, especially students; the machines are winning over the humans by a long way. However the dumbness showed by machines to the easiest of things is also in no comparison to the dumbness of humans, even students.

Today the machines can compute all possibilities of everything good and bad against every action and reaction over time, to tell you smart moves, smart enough to flatter people. The machines could interpret the vague, false and incomplete narrations of your friends to present you the truth. The machines could act logically stop you from making decisions, which you may take in a flow and later regret. And what not, it is a pool of possibilities. So the machine package sounds far better than any living companion, only till it comes to a context unseen, unheard, not thought of, unimaginable and a surprise when the acts of your machine package would be mere entertaining rather than helping.

Theory Syllabus

S. No. Topic Details
1. Introduction Definition, Foundations, History, Current AI systems
2. Intelligent Agents Agents and environment, Rationality, PEAS, Nature of Environment, Different types of agents
3. Searching Agent design, Toy Problems, Searching, Tree Search and Graph Search, Uninformed Search, Breadth First Search, Depth First Search, Depth-Limited Search, Iterative Deepening, Iterative Lengthening, Bidirectional Search, Sensorless problems, Contingency problems.
4. Informed Search Informed/Heuristic Search, Heuristic Search, A* Search, Memory bounded heuristic search, heuristic functions, local search and optimization, hill-climbing, simulated annealing, local beam search, online search, online depth first search.
5. Constraint Satisfaction Problems Constraint Satisfaction Problems, Backtracking, Minimum Remaining Values heuristic, Most Constraint Variable heuristic, Least Constraining Value heuristic, Forward Checking, Constraint Propagation, local search, problem decomposition.
6. Adversarial Search Games, optimal decisions in games, minimax algorithm, multi-player games, alpha-beta pruning, evaluation functions, cutting off search, expectiminimax algorithm, dice/card games.  
7. Planning The planning problem, language specification and PDDL, examples of planning problems, forward search, backward search, heuristics, partial order planning, planning graphs, heuristics from planning graphs, Graphplan algorithm.
8. Uncertainty Uncertainty, probability basics, axioms of probability, inference using full joint distributions, independence, Bayes' rule, Naive Bayes.  
9. Probabilistic Reasoning Representation, Bayesian Networks, Construction of Bayesian Networks, Conditional Independence, Bayesian Networks with continuous variables.
10. Making Simple Decisions Beliefs, Desires and Uncertainty, Utility Theory, Utility Functions, Multiattribute Utility Functions, Decision Networks, Value of Information
11. Making Complex Decisions Stochastic Problems, Value Iteration, Policy Iteration, Game Theory
12. Reinforcement Learning Reinforcement Learning, Passive Reinforcement Learning, Direct utility estimation, Active Dynamic Programming, Temporal Difference Learning, Active Reinforcement Learning, Exploration and Exploitation, Q-Learning.

Lab Syllabus

S. No. Topic
1. Design of a simple reflex agent
2, Searching using Breadth First Search
3. Searching using Uniform Cost Search
4. Searching using A* Algorithm and Heuristic Search
5. Constraint Satisfaction Problems using Minimum Remaining Values, Most Constrained Variable, Least Constraining Value Heuristics
6. Adversarial Search
7. Value Iteration
8. Policy Iteration

Pre-Requisites

Willingness to look at real world probelms and design methodologies for solving them will be very helpful for the course. Good algorithmic and programming skills are desirable.

Text/Reference Material

Myself on:

And also on:

Contact

Dr. Rahul Kala
Assistant Professor,
IIIT Allahabad,

Phone: +91 532 299 2117
Mobile: +91 7054 292 063
E-mail: rkala@iiita.ac.in, rkala001@gmail.com