Seminar: Advanced Topics in
Reinforcement Learning and Planning
Summer 2023

Hector Geffner (i6; RLeap), Computer Science, RWTH



Type of course: Seminar

Study programs:
Master Computer Science
Master Data Science
Master Software Systems Engineering


Offering chair: 
Machine Learning and Reasoning (i6), RWTH

Description: Students will present papers from a list of important, recent works in machine learning compiled by the professor. Topics include deep and reinforcement learning, and transformer and GNN architectures. The focus will be in the use of these techniques in  simple applications that are crisp and easy to understand. The list of papers will also include those underlying  well-known learning systems like  as ChaptGPT, AlphaZero, Gato, etc. Final list of papers and topics to be covered to be defined after the first meeting.

 This is a new seminar by a new professor and the offer is coming a bit late. Applications will be accepted until the kickoff meeting.


Recommended prior knowledge: Bachelor degree in CS or Equivalent. Basic AI and ML courses.

Kickoff: 28/3/2023, 15h-17h, Room 228, Theaterstr. 35, 2nd floor

Format, organization, evaluation: TBA

Website: https://www-i6.informatik.rwth-aachen.de/~hector.geffner/Seminar-S2023.html

Introduction -- Slides 

Tentative, partial list of papers


RL/Planning


Playing Atari with Deep Reinforcement Learning
Volodymyr Mnih, Koray Kavukcuoglu, David Silver, et al, 2013
 
Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm
David Silver, Thomas Hubert, Julian Schrittwieser, et al, 2017

Mastering Atari, Go, Chess and Shogi by Planning with a Learned Model
Julian Schrittwieser, Ioannis Antonoglou, Thomas Hubert, et al, 2020

Transformers are Sample-Efficient World Models
Vincent Micheli, Eloi Alonso, Francois Fleuret, 2022

A Generalist Agent
Scott Reed, Konrad Zolna, Emilio Parisotto, et al., 2023


Deep learning: Transformers

End-to-end symbolic regression with transformers
Pierre-Alexandre Kamienny, Stephane d'Ascoli, Guillaume Lample, Francois Charton, 2022
 
Linear algebra with transformers
Francois Charton, 2021
 
What Can Transformers Learn In-Context? A Case Study of Simple Function Classes
Shivam Garg, Dimitris Tsipras, Percy Liang, Gregory Valiant, 2023
 


Large Language Models, RL, Planning


Training language models to follow instructions with human feedback (ChatGPT)
Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, et al, 2022

ReAct: Synergizing Reasoning and Acting in Language Models
S.  Yao, J. Zhao, D. Yu, et al. 3/2023

Guiding Pretraining in Reinforcement Learning with Large Language Models.
Yuqing Du, Olivia Watkins, Zihan Wang, Cédric Colas, Trevor Darrell, Pieter Abbeel, Abhishek Gupta, Jacob Andreas. 2/2023

On the Planning Abilities of Large Language Models (A Critical Investigation with a Proposed Benchmark)
K.  Valmeekam, S.  Sreedharan, M.  Marquez, A.  Olmo, S. Kambhampati. 2/2023