Fictional Co-Play for Human-Agent Collaboration

Fictional Co-Play for Human-Agent Collaboration: Evaluating state-of-the-art reinforcement learning technique for adaptability to human collaborators

Ordonez Cardenas, Nathan (TU Delft Electrical Engineering, Mathematics and Computer Science)

Oliehoek, F.A. (mentor)
Loftin, R.T. (graduation committee)

Delft University of Technology

Computer Science and Engineering

CSE3000 Research Project

2022-06-27

A longstanding problem in the area of reinforcement learning is human-agent col- laboration. As past research indicates that RL agents undergo a distributional shift when they start collaborating with human beings, the goal is to create agents that can adapt. We build upon research using the two-player Overcooked environment to repro- duce a simplified version of the Fictitious Co-Play algorithm in order to confirm past found improvements at a smaller scale of training and using Self-Play and Population- based trained algorithms as the baselines for comparison. We find that the agent on average slightly outperforms both baseline algorithms when evaluated using a human proxy. We also find high cross-seed variance in performance, indicating the potential for further hyperparameter tuning.

Reinforcement Learning
Ad-hoc teamwork
human-ai collaboration
Human-Agent Teamwork
Overcooked AI

http://resolver.tudelft.nl/uuid:ca39cc40-049a-42ce-ba6f-003e5c358351

Student theses

bachelor thesis

© 2022 Nathan Ordonez Cardenas