中文
相关论文

相关论文: Reward Poisoning in Reinforcement Learning: Attack…

200 篇论文

Federated learning (FL) has become a popular tool for solving traditional Reinforcement Learning (RL) tasks. The multi-agent structure addresses the major concern of data-hungry in traditional RL, while the federated mechanism protects the…

机器学习 · 计算机科学 2024-01-08 Evelyn Ma , Praneet Rathi , S. Rasoul Etesami

Adversarial samples are perturbed inputs crafted to mislead the machine learning systems. A training mechanism, called adversarial training, which presents adversarial samples along with clean samples has been introduced to learn robust…

计算机视觉与模式识别 · 计算机科学 2018-08-07 Vivek B. S. , Konda Reddy Mopuri , R. Venkatesh Babu

We introduce Random Reward Perturbation (RRP), a novel exploration strategy for reinforcement learning (RL). Our theoretical analyses demonstrate that adding zero-mean noise to environmental rewards effectively enhances policy diversity…

机器学习 · 计算机科学 2025-06-11 Haozhe Ma , Guoji Fu , Zhengding Luo , Jiele Wu , Tze-Yun Leong

Much of the current work on reinforcement learning studies episodic settings, where the agent is reset between trials to an initial state distribution, often with well-shaped reward functions. Non-episodic settings, where the agent must…

机器学习 · 计算机科学 2020-06-23 John D. Co-Reyes , Suvansh Sanjeev , Glen Berseth , Abhishek Gupta , Sergey Levine

Safe reinforcement learning is a promising path toward applying reinforcement learning algorithms to real-world problems, where suboptimal behaviors may lead to actual negative consequences. In this work, we focus on the setting where…

机器学习 · 计算机科学 2022-02-17 Garrett Thomas , Yuping Luo , Tengyu Ma

Manipulating the interaction trajectories between the intelligent agent and the environment can control the agent's training and behavior, exposing the potential vulnerabilities of reinforcement learning (RL). For example, in Cyber-Physical…

机器学习 · 计算机科学 2024-11-21 Zhi Luo , Xiyuan Yang , Pan Zhou , Di Wang

Attacks on machine learning models have been extensively studied through stateless optimization. In this paper, we demonstrate how a reinforcement learning (RL) agent can learn a new class of attack algorithms that generate adversarial…

密码学与安全 · 计算机科学 2025-11-20 Kyle Domico , Jean-Charles Noirot Ferrand , Ryan Sheatsley , Eric Pauley , Josiah Hanna , Patrick McDaniel

Recent studies have shown that deep reinforcement learning agents are vulnerable to small adversarial perturbations on the agent's inputs, which raises concerns about deploying such agents in the real world. To address this issue, we…

机器学习 · 计算机科学 2021-11-12 Tuomas Oikarinen , Wang Zhang , Alexandre Megretski , Luca Daniel , Tsui-Wei Weng

In recent years, the security issues of artificial intelligence have become increasingly prominent due to the rapid development of deep learning research and applications. Backdoor attack is an attack targeting the vulnerability of deep…

密码学与安全 · 计算机科学 2023-12-14 Peixin Zhang , Jun Sun , Mingtian Tan , Xinyu Wang

This study investigates a counterintuitive phenomenon in adversarial machine learning: the potential for noise-based defenses to inadvertently aid evasion attacks in certain scenarios. While randomness is often employed as a defensive…

密码学与安全 · 计算机科学 2024-11-01 Steve Bakos , Pooria Madani , Heidar Davoudi

In practical applications, we can rarely assume full observability of a system's environment, despite such knowledge being important for determining a reactive control system's precise interaction with its environment. Therefore, we propose…

机器学习 · 计算机科学 2022-06-24 Edi Muskardin , Martin Tappler , Bernhard K. Aichernig , Ingo Pill

Reinforcement Learning (RL) is a promising approach for achieving autonomous driving due to robust decision-making capabilities. RL learns a driving policy through trial and error in traffic scenarios, guided by a reward function that…

机器人学 · 计算机科学 2026-03-06 Ahmed Abouelazm , Jonas Michel , Helen Gremmelmaier , Tim Joseph , Philip Schörner , J. Marius Zöllner

Adversarial training in reinforcement learning (RL) is challenging because perturbations cascade through trajectories and compound over time, making fixed-strength attacks either overly destructive or too conservative. We propose…

机器学习 · 计算机科学 2026-01-30 Lucas Schott , Elies Gherbi , Hatem Hajri , Sylvain Lamprier

Reinforcement learning (RL) -- algorithms that teach artificial agents to interact with environments by maximising reward signals -- has achieved significant success in recent years. These successes have been facilitated by advances in…

机器学习 · 计算机科学 2025-04-03 Llewyn Salt , Marcus Gallagher

In recent years, unmanned aerial vehicle (UAV) related technology has expanded knowledge in the area, bringing to light new problems and challenges that require solutions. Furthermore, because the technology allows processes usually carried…

人工智能 · 计算机科学 2022-12-08 Cristian Millán-Arias , Ruben Contreras , Francisco Cruz , Bruno Fernandes

The desire to make applications and machines more intelligent and the aspiration to enable their operation without human interaction have been driving innovations in neural networks, deep learning, and other machine learning techniques.…

机器学习 · 计算机科学 2022-09-30 Fadi AlMahamid , Katarina Grolinger

Reinforcement learning (RL), particularly in sparse reward settings, often requires prohibitively large numbers of interactions with the environment, thereby limiting its applicability to complex problems. To address this, several prior…

机器学习 · 计算机科学 2020-11-20 Prasoon Goyal , Scott Niekum , Raymond J. Mooney

Reinforcement learning (RL) systems typically optimize scalar reward functions that assume precise and reliable evaluation of outcomes. However, real-world objectives--especially those derived from human preferences--are often uncertain,…

机器学习 · 计算机科学 2026-04-30 Disha Singha

While reinforcement learning (RL) has the potential to enable robots to autonomously acquire a wide range of skills, in practice, RL usually requires manual, per-task engineering of reward functions, especially in real world settings where…

机器人学 · 计算机科学 2019-02-15 Tianhe Yu , Gleb Shevchuk , Dorsa Sadigh , Chelsea Finn

Empowering safe exploration of reinforcement learning (RL) agents during training is a critical challenge towards their deployment in many real-world scenarios. When prior knowledge of the domain or task is unavailable, training RL agents…

人工智能 · 计算机科学 2025-08-27 Daniel Bethell , Simos Gerasimou , Radu Calinescu , Calum Imrie