Reproduce “Finite-Memory Control of POMDPs: Fundamental Limits and Efficient Design” 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
Emrecan Kutay, Atilla Eryilmaz, Ness Shroff
RECAP is a finite-memory controller for partially observable Markov decision processes. It recursively updates compressed hidden-state predictions, preserving belief distinctions that can change actions when observations do not resolve them. The authors derive a lower bound on losses from insufficient memory. Across 11 benchmarks, RECAP leads or ties budget-matched methods on nine and ranks second on two, establishing results only within this set.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.