Differentiable Trust Region Projection Layers

F Otto - 2024‏ - ub01.uni-tuebingen.de
Deep reinforcement learning and especially policy gradient methods have achieved
remarkable success in various domains. However, challenges remain for policy gradient …

[PDF][PDF] Digitalized Energy Systems Carl von Ossietzky Universität Oldenburg Ammerländer Heerstraße 114-118, 26129 Oldenburg, thomas. wolgast@ uni-oldenburg …

T Wolgast‏ - researchgate.net
ABSTRACT The design of Reinforcement Learning (RL) environments has a strong impact
on RL training performance and generality of results. While most researchers focus on the …

Detecting danger in gridworlds using Gromov's Link Condition

TF Burns, R Tang - arxiv preprint arxiv:2201.06274, 2022‏ - arxiv.org
Gridworlds have been long-utilised in AI research, particularly in reinforcement learning, as
they provide simple yet scalable models for many real-world applications such as robot …