COL333/COL671/COL7333/AIL7025 - Artificial Intelligence - Autumn 2025
Tuesday, Thursday, Friday 11-11:50 pm in LH 108


Instructor: Mausam
(mausam at cse dot iitd dot ac dot in)
Office hours: by appointment, SIT 402
TAs: TBD (office hours by appointment)

Course Contents

Introduction; philosophy of intelligent agents; uninformed search; heuristic search; local search; constraint satisfaction; logic and satisfiability; adversarial search; decision theory; Markov decision processes; Bayesian networks representation, inference and learning; reinforcement learning; basics of supervised, semi-supervised and unsupervised learning; intro to deep learning; intro to NLP; current research trends.

Schedule

Start End Topics & Lecture Notes Required Readings Additional Resources
Jul 24 Jul 24 [Slides] Administrivia
[Video] What to Expect from Course


Aug 5Aug 14 [Slides] Introduction
[Video] What to Expect from AI
[Video] History of AI from 40s - 90s
[Video] History of AI in the 90s
[Video] History of AI in NASA & DARPA (2000s)
[Video] The Present State of AI
[Video] Definition of AI (Dictionary Meaning)
[Video] Definition of AI (Thinking vs. Acting and Humanly vs. Rationally)
[Video] Definition of AI (Rational Agent View of AI)
[Video] Examples Tasks, Phases of AI & Course Plan
AIMA Chapter 1
Beyond Turing Test
Are we in AI bubble?
Benefits/Risks of AI
Introduction to AI: Past, Present & Future
Jul 25Aug 5 [Slides] Uninformed search
[Video] Uninformed Search: Notion of a State
[Video] Uninformed Search: Search Problem and Examples
[Video] Uninformed Search: Basic Search Strategies
[Video] Uninformed Search: Iterative Deepening DFS
[Video] Uninformed Search: Bidirectional Search
AIMA Chapter 3.1-3.4
Beam Search
Search Algorithms Performance
Uniform Cost Search vs. Djikstra's
Adventures in State Space
Aug 19 Aug 22 [Slides] Informed search
[Video] Informed Search: Best First Search
[Video] Informed Search: Greedy Best First Search and A* Search
[Video] Informed Search: Analysis of A* Algorithm
[Video] Informed Search: Proof of optimality of A*
[Video] Informed Search: Iterative Deepening A* and Depth First Branch & Bound
[Video] Informed Search: Admissible Heuristics and Domain Relaxation
[Video] Informed Search: Pattern Database Heuristics
AIMA Chapter 3.5-3.7
IDA*
Depth First Branch and Bound
The Beauty of A*
A*/IDA* Example
Aug 26 Aug 29 [Slides] Local search
[Video] Local Search: Satisfaction vs. Optimization
[Video] Local Search: The Example of N-Queens
[Video] Local Search: Hill Climbing
[Video] Local Search: Drawbacks of Hill Climbing
[Video] Local Search: Hill Climbing with Random Walk & Random Restart
[Video] Local Search: Simulated Anealing
[Video] Local Search: Local Beam Search and Genetic Algorithms
AIMA Chapter 4.1
Stochastic Beam Search
Evolving Monalisa through Genetic Algorithms
Evolving TSP with Genetic Algorithms
Mixability for Genetic Algorithms (pages 66-68)
Aug 25 Sep 8 Programming Assignment 1
Resources
Sep 2Sep 4 [Slides] Adversarial search
[Video] Adversarial Search: Minimax Algorithm for two player games
[Video] Adversarial Search: An Example of Minimax Search
[Video] Adversarial Search: Alpha Beta Pruning
[Video] Adversarial Search: Analysis of Alpha Beta Pruning
[Video] Adversarial Search: Cutting Off Search
[Video] Adversarial Search: Horizon Effect, Game Databases & Other Ideas
[Video] Adversarial Search: Summary and Other Games
AIMA Chapter 5
How Intelligent is Deep Blue?
Minimax Applet
Sep 9 Sep 12 [Slides] Constraint Satisfaction
[Video] Constraint Satisfaction: Representation of the atomic state
[Video] Constraint Satisfaction: Map coloring and other examples of CSP
[Video] Constraint Satisfaction: Backtracking Search
[Video] Constraint Satisfaction: Variable and Value Ordering in Backtracking Search
[Video] Constraint Satisfaction: Inference for detecting failures early
[Video] Constraint Satisfaction: Exploiting problem structure
AIMA Chapter 6 (skip 6.3.3)
Conversion to Binary CSP
NumberJack
Constraint Programming
Sep 10 Sep 26 Programming Assignment 2
Resources
Sep 18Sep 25 [Slides] Logic and Satisfiability
[Slides] Backdoors in Satisfiability
[Video] Logic in AI: Different Knowledge Representation Systems
[Video] Logic in AI: Syntax
[Video] Logic in AI: Semantics
[Video] Logic in AI: Forward Chaining
[Video] Logic in AI: Resolution
[Video] Logic in AI: Reduction to Satisfiability Problems
[Video] Logic in AI: SAT Solvers - DPLL Algorithm
[Video] Logic in AI: Sat Solvers - WalkSAT Algorithm
AIMA Chapter 7
Advanced SAT Solvers (Sections 2.3, 2.4)
Backdoors
Oct 3 Oct 15 Programming Assignment 3
Code   README
Oct 7 Oct 10 [Slides] Intro to Probability
[Slides] Bayesian Networks Representation
[Video] Uncertainty in AI: Motivation
[Video] Uncertainty in AI: Basics of Probability
[Video] Uncertainty in AI: Conditional Independence & Bayes Rule
[Video] Bayesian Networks: Syntax
[Video] Bayesian Networks: Factorization
[Video] Bayesian Networks: Conditional Independences and d-Separation
[Video] Bayesian Networks: Inference using Variable Elimination
AIMA Chapter 12
AIMA Chapter 13.1-13.2
History of Bayes Theorem
Influence Flow in Bayes Nets
Oct 14 Oct 17 [Slides] Approximate Inference in Bayesian Networks
[Slides] Learning in Bayesian Networks
[Video] Bayesian Networks: Reducing 3-SAT to Bayes Net
[Video] Bayesian Networks: Rejection Sampling
[Video] Bayesian Networks: Likelihood Weighting
[Video] Bayesian Networks: MCMC with Gibbs Sampling
[Video] Bayesian Networks: Maximum Likelihood Learning
[Video] Bayesian Networks: Maximum a-Posteriori Learning
[Video] Bayesian Networks: Bayesian Learning
[Video] Bayesian Networks: Structure Learning and Expectation Maximization
AIMA Chapter 13.3-13.4
AIMA Chapter 20
Log Probabilities
Oct 17 Oct 31 Programming Assignment 4 Resources
Oct 21 Oct 24 [Slides] Decision Theory
[Slides] Markov Decision Processes
[Video] Decision Theory: Steps in Decision Theory
[Video] Decision Theory: Non Deterministic Uncertainty
[Video] Decision Theory: Probabilistic Uncertainty & Value of perfect information
[Video] Decision Theory: Expected Utility vs Expected Value
[Video] Markov Decision Processes: Definition
[Video] Markov Decision Processes: An Example of a Policy
[Video] Markov Decision Processes: Policy Evaluation using System of Linear Equations
[Video] Markov Decision Processes: Iterative Policy Evaluation
[Video] Markov Decision Processes: Value Iteration
[Video] Markov Decision Processes: Policy Iteration and Applications & Extensions of MDPs
AIMA Chapter 16.1-16.3, 16.6, 17.1-17.2.2

Oct 28Oct 31 [Slides] Reinforcement Learning
[Video] Reinforcement Learning: Background
[Video] Reinforcement Learning: Model-based Learning for policy evaluation (Passive Learning)
[Video] Reinforcement Learning: Model-free Learning for policy evaluation (Passive Learning)
[Video] Reinforcement Learning: TD Learning
[Video] Reinforcement Learning: TD Learning and Computational Neuroscience
[Video] Reinforcement Learning: Q Learning
[Video] Reinforcement Learning: Exploration vs Exploitation Tradeoff
[Video] Reinforcement Learning: Generalization in RL
AIMA Chapter 22.1-22.4.2
TD Learning for Backgammon
Oct 31 Nov 14 Programming Assignment 5

Nov 7Nov 13 [Slides] Introduction to Deep Learning and CNNs
[Slides] Deep Reinforcement Learning
[Video] Deep Learning: Perceptrons and Activation functions
[Video] Deep Learning: Example of Handwritten digit recognition
[Video] Deep Learning: Neural Layer as matrix operations
[Video] Deep Learning: Differentiable loss function
[Video] Deep Learning: Backpropagation through a computational graph
[Video] Deep Learning: Thin Deep Vs Fat Shallow Networks
[Video] Deep Learning: Convolutional Neural Networks
[Video] Deep Learning: Deep Reinforcement Learning
AIMA Chapter 21.1-21.3, 21.4.1


Textbooks

Stuart Russell & Peter Norvig, Artificial Intelligence: A Modern Approach,
Prentice-Hall, Fourth Edition (2021) (required).

Ina GoodFellow, Yoshua Bengio & Aaron Courville, Deep Learning,
MIT Press (2016).

Grading

Assignments: 40%; Midterm: 18%; Final: 27%; Quizzes: 15%; Class participation, online discussions: extra credit.

There will be five programming assignments due approximately every two weeks.

Course Administration and Policies

Cheating Vs. Collaborating Guidelines

As adapted from Dan Weld's guidelines.

Collaboration is a very good thing. On the other hand, cheating is considered a very serious offense. Please don't do it! Concern about cheating creates an unpleasant environment for everyone. If you cheat, you get a zero in the assignment, and additionally you risk losing your position as a student in the department and the institute. The department's policy on cheating is to report any cases to the disciplinary committee. What follows afterwards is not fun.

So how do you draw the line between collaboration and cheating? Here's a reasonable set of ground rules. Failure to understand and follow these rules will constitute cheating, and will be dealt with as per institute guidelines.

LLM Policy

We treat an LLM as another student taking the class, and/or a book that you can read. That is, it is OK to seek help from an LLM, while following all the rules above (the RTI Rule, and KSBKBT Rule). But you cannot copy any piece of code or document from an LLM. For example, interact with an LLM to get intuition about the assignment, critique ideas, and brainstorm. Also, use an LLM to learn syntax for a language, get help with debugging. But do NOT use an LLM to write code that you copy/paste in the assignments. Instead, write your own code yourselves.