Time-Constrained Robust MDPs

A Zouitine, D Bertoin, P Clavier… - Advances in Neural …, 2025 - proceedings.neurips.cc
Robust reinforcement learning is essential for deploying reinforcement learning algorithms
in real-world scenarios where environmental uncertainty predominates. Traditional robust …

Rrls: Robust reinforcement learning suite

A Zouitine, D Bertoin, P Clavier, M Geist… - arxiv preprint arxiv …, 2024 - arxiv.org
Robust reinforcement learning is the problem of learning control policies that provide
optimal worst-case performance against a span of adversarial environments. It is a crucial …

Solving robust MDPs as a sequence of static RL problems

A Zouitine, M Geist, E Rachelson - arxiv preprint arxiv:2410.06212, 2024 - arxiv.org
Designing control policies whose performance level is guaranteed to remain above a given
threshold in a span of environments is a critical feature for the adoption of reinforcement …

Revisiting the static model in robust reinforcement learning

A Zouitine, M Geist, E Rachelson - 2023 - openreview.net
Designing control policies whose performance level is guaranteed to remain above a given
threshold in a span of environments is a critical feature for the adoption of reinforcement …

Robust Reinforcement Learning: Theory and Practice

P Clavier - 2024 - theses.hal.science
Reinforcement learning (RL) is a machine learning paradigm that addresses the issue of
sequential decision-making. In this paradigm, the algorithm, designated as an agent …