English
Related papers

Related papers: Provably Efficient Black-Box Action Poisoning Atta…

200 papers

Collecting training data from untrusted sources exposes machine learning services to poisoning adversaries, who maliciously manipulate training data to degrade the model accuracy. When trained on offline datasets, poisoning adversaries have…

Machine Learning · Computer Science 2021-10-27 Tianyu Pang , Xiao Yang , Yinpeng Dong , Hang Su , Jun Zhu

Evaluating deep reinforcement learning (DRL) agents against targeted behavior attacks is critical for assessing their robustness. These attacks aim to manipulate the victim into specific behaviors that align with the attacker's objectives,…

Machine Learning · Computer Science 2024-12-17 Fengshuo Bai , Runze Liu , Yali Du , Ying Wen , Yaodong Yang

Deep Reinforcement Learning (RL) agents are susceptible to adversarial noise in their observations that can mislead their policies and decrease their performance. However, an adversary may be interested not only in decreasing the reward,…

Machine Learning · Computer Science 2022-12-13 Dennis Gross , Thiago D. Simao , Nils Jansen , Guillermo A. Perez

Backdoor attacks manipulate model predictions by inserting innocuous triggers into training and test data. We focus on more realistic and more challenging clean-label attacks where the adversarial training examples are correctly labeled.…

Machine Learning · Computer Science 2023-10-31 Wencong You , Zayd Hammoudeh , Daniel Lowd

Adversarial attacks based on randomized search schemes have obtained state-of-the-art results in black-box robustness evaluation recently. However, as we demonstrate in this work, their efficiency in different query budget regimes depends…

Machine Learning · Computer Science 2021-11-23 Maksym Yatsura , Jan Hendrik Metzen , Matthias Hein

The open-endedness of large language models (LLMs) combined with their impressive capabilities may lead to new safety issues when being exploited for malicious use. While recent studies primarily focus on probing toxic outputs that can be…

Computation and Language · Computer Science 2023-11-30 Jiaxin Wen , Pei Ke , Hao Sun , Zhexin Zhang , Chengfei Li , Jinfeng Bai , Minlie Huang

Deep Learning has become popular due to its vast applications in almost all domains. However, models trained using deep learning are prone to failure for adversarial samples and carry a considerable risk in sensitive applications. Most of…

Computer Vision and Pattern Recognition · Computer Science 2024-05-13 Satyadwyoom Kumar , Saurabh Gupta , Arun Balaji Buduru

Research in adversarial machine learning (AML) has shown that statistical models are vulnerable to maliciously altered data. However, despite advances in Bayesian machine learning models, most AML research remains concentrated on classical…

Machine Learning · Statistics 2025-06-04 Matthieu Carreau , Roi Naveiro , William N. Caballero

Ensuring safety for black-box hybrid dynamical systems presents significant challenges due to their instantaneous state jumps and unknown explicit nonlinear dynamics. Existing solutions for strict safety constraint satisfaction, like…

Robotics · Computer Science 2026-04-27 Aayushi Shrivastava , Kartik Nagpal , Sairam Jinkala , Jean-Baptiste Bouvier , Negar Mehr

Recent work has shown how easily white-box adversarial attacks can be applied to state-of-the-art image classifiers. However, real-life scenarios resemble more the black-box adversarial conditions, lacking transparency and usually imposing…

Cryptography and Security · Computer Science 2021-07-14 Andrei Ilie , Marius Popescu , Alin Stefanescu

Reinforcement learning (RL) has demonstrated its ability to solve high dimensional tasks by leveraging non-linear function approximators. However, these successes are mostly achieved by 'black-box' policies in simulated domains. When…

Machine Learning · Computer Science 2021-11-19 Riad Akrour , Davide Tateo , Jan Peters

Large Language Model (LLM) agents can leverage tools such as Google Search to complete complex tasks. However, this tool usage introduces the risk of indirect prompt injections, where malicious instructions hidden in tool outputs can…

Machine Learning · Computer Science 2025-10-08 Zizhao Wang , Dingcheng Li , Vaishakh Keshava , Phillip Wallis , Ananth Balashankar , Peter Stone , Lukas Rutishauser

Agentic reinforcement learning (RL) trains large language models to autonomously call tools during reasoning, with search as the most common application. These models excel at multi-step reasoning tasks, but their safety properties are not…

Computation and Language · Computer Science 2025-10-21 Yushi Yang , Shreyansh Padarha , Andrew Lee , Adam Mahdi

A reinforcement learning (RL) control policy could fail in a new/perturbed environment that is different from the training environment, due to the presence of dynamic variations. For controlling systems with continuous state and action…

Robotics · Computer Science 2022-08-31 Y. Cheng , P. Zhao , F. Wang , D. J. Block , N. Hovakimyan

Collaborative multi-agent reinforcement learning has rapidly evolved, offering state-of-the-art algorithms for real-world applications, including sensitive domains. However, a key challenge to its widespread adoption is the lack of a…

Machine Learning · Computer Science 2026-01-22 Amine Andam , Jamal Bentahar , Mustapha Hedabou

As reinforcement learning agents become increasingly deployed in real-world scenarios, predicting future agent actions and events during deployment is important for facilitating better human-agent interaction and preventing catastrophic…

Artificial Intelligence · Computer Science 2024-10-31 Stephen Chung , Scott Niekum , David Krueger

Deep reinforcement learning has shown promising results in learning control policies for complex sequential decision-making tasks. However, these neural network-based policies are known to be vulnerable to adversarial examples. This…

Computer Vision and Pattern Recognition · Computer Science 2017-10-04 Yen-Chen Lin , Ming-Yu Liu , Min Sun , Jia-Bin Huang

Machine Learning (ML) algorithms are vulnerable to poisoning attacks, where a fraction of the training data is manipulated to deliberately degrade the algorithms' performance. Optimal poisoning attacks, which can be formulated as bilevel…

Machine Learning · Computer Science 2020-06-23 Javier Carnerero-Cano , Luis Muñoz-González , Phillippa Spencer , Emil C. Lupu

Offline Reinforcement Learning from Human Feedback (RLHF) pipelines such as Direct Preference Optimization (DPO) train on a pre-collected preference dataset, which makes them vulnerable to preference poisoning attack. We study label flip…

Machine Learning · Computer Science 2026-05-26 Chenye Yang , Weiyu Xu , Lifeng Lai

A policy is said to be robust if it maximizes the reward while considering a bad, or even adversarial, model. In this work we formalize two new criteria of robustness to action uncertainty. Specifically, we consider two scenarios in which…

Machine Learning · Computer Science 2019-05-08 Chen Tessler , Yonathan Efroni , Shie Mannor