paper-with-me

Papers

Towards Robust Model-Based Reinforcement Learning Against Adversarial Corruption

2024-02-14 · Chenlu Ye, Jiafan He, Quanquan Gu, Tong Zhang

This study tackles the challenges of adversarial corruption in model-based reinforcement learning (RL), where the transition dynamics can be corrupted by an adversary. Existing studies on corruption-robust RL mostly focus on the setting of model-free RL, where robust least-square regression is often employed for value function estimation. However, these techniques cannot be directly applied to model-based RL. In this paper, we focus on model-based RL and take the maximum likelihood estimation (MLE) approach to learn transition model. Our work encompasses both online and offline settings. In the online setting, we introduce an algorithm called corruption-robust optimistic MLE (CR-OMLE), which leverages total-variation (TV)-based information ratios as uncertainty weights for MLE. We prove that CR-OMLE achieves a regret of $\tilde{\mathcal{O}}(\sqrt{T} + C)$, where $C$ denotes the cumulative corruption level after $T$ episodes. We also prove a lower bound to show that the additive dependence on $C$ is optimal. We extend our weighting technique to the offline setting, and propose an algorithm named corruption-robust pessimistic MLE (CR-PMLE). Under a uniform coverage condition, CR-PMLE exhibits suboptimality worsened by $\mathcal{O}(C/n)$, nearly matching the lower bound. To the best of our knowledge, this is the first work on corruption-robust model-based RL algorithms with provable guarantees.

📄 PDF Abstract BibTeX arXiv:2402.08991

Code (0)

등록된 구현이 없습니다.

Tasks

Model-based Reinforcement Learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)

Methods 이 논문이 사용한 방법론

Focus 설명 없음

Similar Papers 제목 키워드 기반

Robustness of SAM: Segment Anything Under Corruptions and Beyond

2023-06-13 · Yu Qiao, Chaoning Zhang, Taegoo Kang, Donghun Kim 외

Segment anything model (SAM), as the name suggests, is claimed to be capable of cutting out any object and demonstrates impressive zero-shot transfer performance with the guidance of prompts. However, there is currently …

Style Transfer

Defending Against Image Corruptions Through Adversarial Augmentations

2021-04-02 · ICLR 2022 4 · Dan A. Calian, Florian Stimberg, Olivia Wiles, Sylvestre-Alvise Rebuffi 외

Modern neural networks excel at image classification, yet they remain vulnerable to common image corruptions such as blur, speckle noise or fog. Recent methods that focus on this problem, such as AugMix and DeepAugment, …

image-classificationImage Classification

On the effectiveness of adversarial training against common corruptions

2021-03-03 · Klim Kireev, Maksym Andriushchenko, Nicolas Flammarion

The literature on robustness towards common corruptions shows no consensus on whether adversarial training can improve the performance in this setting. First, we show that, when used with an appropriately selected pertur…

Data Augmentation

Exploiting Frequency Spectrum of Adversarial Images for General Robustness

2023-05-15 · Chun Yang Tan, Kazuhiko Kawamoto, Hiroshi Kera

In recent years, there has been growing concern over the vulnerability of convolutional neural networks (CNNs) to image perturbations. However, achieving general robustness against different types of perturbations remain…

Data Augmentation

Detecting Worst-case Corruptions via Loss Landscape Curvature in Deep Reinforcement Learning

2021-09-29 · Ezgi Korkmaz, Jonah Brown-Cohen

The non-robustness of neural network policies to adversarial examples poses a challenge for deep reinforcement learning. One natural approach to mitigate the impact of adversarial examples is to develop methods to detect…

Deep Reinforcement Learningreinforcement-learningReinforcement Learning (RL)