Abstract simulators can be grounded to real tasks by making their dynamics history-dependent and correcting them with real data, enabling RL policy transfer.
Approximate information state for approximate planning and reinforcement learning in partially observed systems,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Finite-memory truncation of infinite-horizon POMGs produces epsilon-Nash equilibria that converge to exact Nash equilibria as truncation length increases under filter stability conditions.
citing papers explorer
-
Abstract Sim2Real through Approximate Information States
Abstract simulators can be grounded to real tasks by making their dynamics history-dependent and correcting them with real data, enabling RL policy transfer.
-
Nash Approximation Gap in Truncated Infinite-horizon Partially Observable Markov Games
Finite-memory truncation of infinite-horizon POMGs produces epsilon-Nash equilibria that converge to exact Nash equilibria as truncation length increases under filter stability conditions.