Reinforcement Learning with Fast and Forgetful Memory

Published: 21 Sept 2023, Last Modified: 13 Jan 2024NeurIPS 2023 posterEveryoneRevisionsBibTeX
Keywords: reinforcement learning, partially observable, POMDP, memory, rnn, transformer
TL;DR: We propose a drop-in alternative to RNNs for model-free recurrent RL
Abstract: Nearly all real world tasks are inherently partially observable, necessitating the use of memory in Reinforcement Learning (RL). Most model-free approaches summarize the trajectory into a latent Markov state using memory models borrowed from Supervised Learning (SL), even though RL tends to exhibit different training and efficiency characteristics. Addressing this discrepancy, we introduce Fast and Forgetful Memory, an algorithm-agnostic memory model designed specifically for RL. Our approach constrains the model search space via strong structural priors inspired by computational psychology. It is a drop-in replacement for recurrent neural networks (RNNs) in recurrent RL algorithms, achieving greater reward than RNNs across various recurrent benchmarks and algorithms _without changing any hyperparameters_. Moreover, Fast and Forgetful Memory exhibits training speeds two orders of magnitude faster than RNNs, attributed to its logarithmic time and linear space complexity. Our implementation is available at
Supplementary Material: zip
Submission Number: 389