Why Linear Recurrent Memory Works in Partially Observable Reinforcement Learning 事件

Why Linear Recurrent Memory Works in Partially Observable Reinforcement Learning arXiv:2605.31261v1 Announce Type: cross Abstract: The family of linear recurrent neural networks has shown strong performance as recurrent memory units in partially observable reinforcement learning. We provide a theoretical justification for their empirical effectiveness by constructing and studying two linear filters: (i) the first exactly reproduces the pre-softmax logits of the belief vector in a hidden Markov