Reproduce “Adapting Behaviour for Learning Progress” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Tom Schaul, Diana Borsa, David Ding, David Szepesvari, Georg Ostrovski, Will Dabney, Simon Osindero
The authors introduce a bandit that adapts an agent’s exploration, sampling behavior settings once per episode and using episode return as a proxy for usefulness. Across 57 Atari games, the adaptive agent reports a median human-normalized score of 113 ± 2%, compared with 118 ± 6% for a fixed reference, supporting comparable performance without per-task tuning; this does not establish a speedup.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.