Reinforcement Learning and Decision Making Under Uncertainty
The course will give a thorough introduction to reinforcement learning. The first 8 weeks will be devoted to the core theory and algorithms of reinforcement learning. The final 6 weeks will be focused on project work, during which more advanced topics will be inroduced.
The first 6 weeks will require the students to complete 3 assignments. The remainder of the term, the students will have to prepare a project, for which they will need to submit a report.
Schedule
- Beliefs and Decisions
Bayes Decision and Game Theory - Bayesian Analysis
Estimation Theory and Concentration Inequalities - Bandit problems
MDPs and belief states - MDP Theory: Value Iteration
Policy Iteration - Temporal Differences
Modified Policy Iteration - Sarsa / Q-Learning
Stochastic Approximation / Actor-Critic - Function Approximation, Gradient Methods
Large-Scale RL - Bayesian RL
- UCB, Regret bounds
UCRL, Regret bounds - Monte Carlo Planning: UCT/MCTS/AlphaZero
- Advanced Bayesian models
- Inverse Reinforcement Learning
- Multiagent extensions: Bayesian Games
- Group work
Details
| Code | 42111 62111 |
| Type | Course |
| ECTS | 5 |
| Site | Neuchâtel |
| Track(s) |
T4 – Theory and Logic T6 – Data Science |
| Semester | S2024 |
Teaching
| Learning Outcomes |
|
| Lecturer(s) |
Christos Dimitrakakis |
| Language | english |
| Course Page | The course page in ILIAS can be found at https://ilias.unibe.ch/goto_ilias3_unibe_crs_2793429.html. |
Schedules and Rooms
| Period | Weekly |
| Schedule | Tuesday, 08:45 - 12:00 |
| Location | UniNE, Unimail |
| Room | B217 |
Additional information
| Comment | First Lecture Evaluation |