- Sat 27 November 2021
- Teaching
- Glen Berseth
- #Reinforcement Learning, #Robot Learning
Course Schedule
Here is an outline of the course calendar.
Week | Date | Topic(s) | Slides | Related paper(s) |
---|---|---|---|---|
1 | d | Behaviour Cloning | d | d |
2 | d | Intro to RL | s | S&B Chapters 2,3 |
2 | d | Policy Gradients | s | TRPO/PPO |
2 | d | Actor-Critic methods | s | A3C |
2 | d | Value function methods | s | Value Iteration and GCG |
2 | d | Q-functions | s | DDPG, TD3, SAC |
2 | d | Optimal Control and planning | s | ? |
2 | d | Model-based RL | s | Pets, MBPO |
2 | d | Exploration | s | ? |
2 | d | Offline RL | s | CQL, |
2 | d | Goal Conditioned RL | s | Goal, Conditioned RL, RIG, DiscoRL |
2 | d | Variational Inference and Generative Models | s | VAE, GAN |
2 | d | The connection between Inference and Control | s | Control as Inference |
2 | d | Inverse Reinforcement Learning | s | MaxEnt IRL, |
2 | d | Hierarchical RL and skill discovery | s | DeepLoco, Hiro, Options, DIYAN |
2 | d | Meta-learning (for RL) | s | ProMP, PEARL |
2 | d | Continual RL | s | PLaid, Survery Paper from Mila |
2 | d | Reward function learning | s | VICE, SOLAR |
2 | d | State abstraction | s | SoRB, RECON |
2 | d | Unsupervised RL | s | SMiRL, IC2 |
2 | d | Generalization in RL | s | Amorpheus, SMP |
2 | d | Multi-Agent RL | s | MADDPG, PoDMPs |