Real-Time Recurrent Learning using Trace Units in Reinforcement Learning

E Elelimy, A White, M Bowling, M White - arxiv preprint arxiv:2409.01449, 2024 - arxiv.org
Recurrent Neural Networks (RNNs) are used to learn representations in partially observable
environments. For agents that learn online and continually interact with the environment, it is …

TOP-ERL: Transformer-based Off-Policy Episodic Reinforcement Learning

G Li, D Tian, H Zhou, X Jiang, R Lioutikov… - arxiv preprint arxiv …, 2024 - arxiv.org
This work introduces Transformer-based Off-Policy Episodic Reinforcement Learning (TOP-
ERL), a novel algorithm that enables off-policy updates in the ERL framework. In ERL …