collaborators
Showing cs.LGShow all

5 papers · 1 filter

cs.LG2026

Preventing Learning Stagnation in PPO by Scaling to 1 Million Parallel Environments

Michael Beukman, Khimya Khetarpal, Zeyu Zheng +4

An agent's performance stagnating at a suboptimal level is a common problem in deep on-policy RL. Focusing on PPO, we show that plateaus in certain regimes arise not because of kno…

cs.LG2025

Beyond Fixed Tasks: Unsupervised Environment Design for Task-Level Pairs

Daniel Furelos-Blanco, Charles Pert, Frederik Kelbel +3

Training general agents to follow complex instructions (tasks) in intricate environments (levels) remains a core challenge in reinforcement learning. Random sampling of task-level…

cs.LG2025

Mitigating Goal Misgeneralization via Minimax Regret

Karim Abdel Sadek, Matthew Farrugia-Roberts, Usman Anwar +4

Safe generalization in reinforcement learning requires not only that a learned policy acts capably in new situations, but also that it uses its capabilities towards the pursuit of…

cs.LG2024

minimax: Efficient Baselines for Autocurricula in JAX

Minqi Jiang, Michael Dennis, Edward Grefenstette +1

Unsupervised environment design (UED) is a form of automatic curriculum learning for training robust decision-making agents to zero-shot transfer into unseen environments. Such aut…

cs.LG2024

Refining Minimax Regret for Unsupervised Environment Design

Michael Beukman, Samuel Coward, Michael Matthews +4

In unsupervised environment design, reinforcement learning agents are trained on environment configurations (levels) generated by an adversary that maximises some objective. Regret…