SUK · Third Year B.Tech. Computer Science and Engineering · NEP 2020
Artificial Intelligence
CSE3101 · Semester V · 3 credits · ISE 10 + mid-semester 30 + ESE 60
Unit 1: Artificial Intelligence and Its Issues
5 hoursDefinitions and importance of AI, its evolution and applications; classification of AI systems by environment; knowledge-inferring systems and planning; uncertainty and learning systems.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
Unit 2: Problem Solving and Heuristic Search
6 hoursProblem solving by search, problem and state spaces; blind search and its types, performance measures; heuristic search; game playing with the minimax algorithm and alpha-beta pruning.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Problem formulation and state-space search
- Uninformed search: BFS, DFS and uniform-cost search
- Informed search: greedy best-first and A*
- Game playing and minimax
- Alpha–beta pruning
Unit 3: Probabilistic Reasoning and Markov Decision Processes
7 hoursProbability, conditional probability and Bayes' rule; Bayesian networks: representation, construction and inference; temporal models and the hidden Markov model; MDP formulation, utility theory and utility functions, value iteration, policy iteration and partially observable MDPs.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Total probability and Bayes' theorem
- Uncertainty and Bayesian networks
- Hidden Markov models
- Finite Markov decision processes
- Value iteration and generalised policy iteration
- Partially observable MDPs
Unit 4: Learning Systems and Expert Systems
7 hoursForms of learning: supervised, unsupervised and reinforcement learning; learning decision trees; expert systems: development stages, probability-based expert systems, tools, difficulties and applications.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Forms of learning and inductive learning
- Learning decision trees
- Expert systems, shells and knowledge acquisition
Unit 5: Reinforcement Learning
5 hoursPassive reinforcement learning, direct utility estimation and adaptive dynamic programming; temporal-difference learning; active reinforcement learning and Q-learning.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Reinforcement learning: elements and scope
- Policy evaluation and policy iteration
- Temporal-difference prediction
- Sarsa and Q-learning
Unit 6: AI with Python
5 hoursPython's libraries for AI: NumPy, SciPy, matplotlib, nltk and Simple AI; installing and running Python; real-time AI applications in Python; case studies of AI platforms: Azure ML, Google AI, Swift AI and TensorFlow.
Examined in the 60-mark end-semester paper, and in the mid-semester exam if it is taught by then
- Python libraries for data analysis
- AI, ML, deep learning and data science
- AI platforms: Azure ML, Google AI and TensorFlow
Unit contents are summarised from the official SUK syllabus . AiCanCode.org is not affiliated with Shivaji University, Kolhapur.