DRATS derives a minimax objective from a feasibility formulation of MTRL to adaptively sample tasks with the largest return gaps, leading to better worst-task performance on MetaWorld benchmarks.
Nature , year=
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
representative citing papers
A diffusion model trained on DOOM play sessions generates stable real-time interactive game frames at 20 FPS with quality near lossy JPEG.
KBSE learns policies and barrier functions iteratively via conditional mean embeddings to bound unsafe state reachability probabilities during exploration in deep RL.
Higher-resolution observations with global-average-pooling encoders improve RL performance and generalization by enabling more localized visual attention, yielding up to 28% gains over standard Impala encoders.
citing papers explorer
-
Distributionally Robust Multi-Task Reinforcement Learning via Adaptive Task Sampling
DRATS derives a minimax objective from a feasibility formulation of MTRL to adaptively sample tasks with the largest return gaps, leading to better worst-task performance on MetaWorld benchmarks.
-
Diffusion Models Are Real-Time Game Engines
A diffusion model trained on DOOM play sessions generates stable real-time interactive game frames at 20 FPS with quality near lossy JPEG.
-
Kernel-Based Safe Exploration in Deep Reinforcement Learning
KBSE learns policies and barrier functions iteratively via conditional mean embeddings to bound unsafe state reachability probabilities during exploration in deep RL.
-
Higher Resolution, Better Generalization: Unlocking Visual Scaling in Deep Reinforcement Learning
Higher-resolution observations with global-average-pooling encoders improve RL performance and generalization by enabling more localized visual attention, yielding up to 28% gains over standard Impala encoders.