RTMNU · Final Year B.Tech Artificial Intelligence & Data Science · CBCS
Reinforcement Learning
BTECHAI&DS703.5T · Elective III · Semester VII · 3 credits · College 30 + university 70
One of the Elective III options.
Unit 1: Introduction to Reinforcement Learning
8 hoursComponents of reinforcement learning, policies and value functions and the types of reinforcement learning; Markov decision processes — agents, environments, discounted rewards, value functions and Bellman equations.
Examined in the 70-mark end-semester paper
- Types of reinforcement learning
- Agent-environment interface
- Markov decision processes
- Rewards, returns and discounting
- Value functions
- Bellman equations
Unit 2: Dynamic Programming and Monte Carlo Methods
7 hoursPolicy and value iteration, generalised policy iteration and extensions of DP; Monte Carlo prediction of value functions and Monte Carlo control, on-policy first-visit and every-visit control.
Examined in the 70-mark end-semester paper
- Dynamic programming for MDPs
- Policy iteration
- Value iteration
- Monte Carlo prediction
- Monte Carlo control
Unit 3: Temporal Difference Learning
8 hoursTD prediction and control (SARSA and Q-learning), eligibility traces, linear and nonlinear function approximation, and deep reinforcement learning.
Examined in the 70-mark end-semester paper
- Temporal difference learning
- SARSA
- Q-learning
- Eligibility traces and TD(λ)
- Value-function approximation
- Semi-gradient methods and linear function approximation
- Deep reinforcement learning
Unit 4: Policy Gradients and Multi-agent RL
7 hoursPolicy approximation and search, gradient estimation and the REINFORCE algorithm; multi-agent reinforcement learning, Markov games, Nash and correlated equilibria, and learning policies for MARL.
Examined in the 70-mark end-semester paper
- Policy gradients: REINFORCE
- Policy gradient theorem
- Multi-agent reinforcement learning
- Markov games, Nash and correlated equilibria
Unit 5: Applications of Reinforcement Learning
6 hoursRobotics, game playing, autonomous driving, recommendation, marketing and advertising, and resource management; the practical relevance of the technology.
Examined in the 70-mark end-semester paper
- Applications of reinforcement learning
- Monte Carlo tree search
Unit contents are summarised from the official RTMNU syllabus . AiCanCode.org is not affiliated with Rashtrasant Tukadoji Maharaj Nagpur University (RTMNU).