Tag
This paper develops a game-theoretic foundation for bisimulation metrics in undiscounted (total- and average-reward) MDPs, interpreting metric operators as Bellman–Shapley game value functions and proving finiteness, aggregation error, and suboptimality guarantees.
该论文将主动推理（Active Inference）与贝叶斯 UCBVI 放在统一框架下比较，发现 AIF 的信息增益衰减过快，需用平方根校准才能达到 √T 遗憾界；校准后的 AIF 在对抗性奖励设定下表现出 ℓ1 置信集优势，而在固定奖励下并不优于 Bayes-UCBVI。
This paper proposes action-conditioned bisimulation over an empirical predictive state graph to decide when GUI agent memories of two web pages should be merged, improving success on MiniWoB++ over memoryless baselines without any training.
This paper proposes a performance-driven state abstraction method for reinforcement learning that directly optimizes decision quality, using a multi-timescale framework to jointly adapt the policy and a tree-structured abstraction. The algorithm refines or aggregates state space based on Q-value discrepancies, achieving better sample efficiency and faster replanning than baselines.
This paper proposes ARS, a memory-augmented agentic recommender system that treats recommendation as a partially observable problem with a hierarchical belief-state memory structure. It achieves state-of-the-art performance on four benchmarks with significant improvements over baselines.