Security of Deep Reinforcement Learning for Autonomous Driving: A Survey

arXiv 2212.06123 ↗First posted 12 Dec 2022cs.LG cs.ROcited by 1

On the map: End-to-end autonomous driving → Adversarial scenarios and safety testing

Abstract (arXiv)

Reinforcement learning (RL) enables agents to learn optimal behaviors through interaction with their environment and has been increasingly deployed in safety-critical applications, including autonomous driving. Despite its promise, RL is susceptible to attacks designed either to compromise policy learning or to induce erroneous decisions by trained agents. Although the literature on RL security has grown rapidly and several surveys exist, existing categorizations often fall short in guiding the selection of appropriate defenses for specific systems. In this work, we present a comprehensive survey of 86 recent studies on RL security, addressing these limitations by systematically categorizing attacks and defenses according to defined threat models and single- versus multi-agent settings. Furthermore, we examine the relevance and applicability of state-of-the-art attacks and defense mechanisms within the context of autonomous driving, providing insights to inform the design of robust RL systems.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Adversarial scenarios and safety testing, one of 9 directions in End-to-end autonomous driving. Builds on 77 of its 121 references that are in the map. First posted on arXiv 12 Dec 2022.

Builds on

77 of 121 references.

Closest ancestors · 2

Same direction, newest first

Other references · 75

2017Proximal Policy Optimization Algorithmscited by 14,441
2015Distilling the Knowledge in a Neural Networkcited by 10,076
2013Playing Atari with Deep Reinforcement Learningcited by 3,211
2016Membership Inference Attacks against Machine Learning Modelscited by 2,221
2017Understanding Black-box Predictions via Influence Functionscited by 1,714
2020D4RL: Datasets for Deep Data-Driven Reinforcement Learningcited by 1,406
2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,052
2016Universal adversarial perturbationscited by 774
2020Deep Reinforcement Learning for Autonomous Driving: A Surveycited by 698
2016Stealing Machine Learning Models via Prediction APIscited by 602
2017Adversarial Patchcited by 514
2016Transferability in Machine Learning: from Phenomena to Black-Box Attacks using Adversarial Samplescited by 439

38 more

2017Robust Adversarial Reinforcement Learningcited by 439
2017Wild Patterns: Ten Years After the Rise of Adversarial Machine Learningcited by 411
2017Adversarial Attacks on Neural Network Policiescited by 314
2018Knockoff Nets: Stealing Functionality of Black-Box Modelscited by 285
2017Adversarial Examples for Semantic Segmentation and Object Detectioncited by 275
2020Robust Deep Reinforcement Learning against Adversarial Perturbations on State Observationscited by 235
2019Adversarial Policies: Attacking Deep Reinforcement Learningcited by 209
2020Self-Distillation Amplifies Regularization in Hilbert Spacecited by 171
2019Action Robust Reinforcement Learning and Applications in Continuous Controlcited by 167
2017Tactics of Adversarial Attack on Deep Reinforcement Learning Agentscited by 163
2018ShapeShifter: Robust Physical Adversarial Attack on Faster R-CNN Object Detectorcited by 159
2017Robust Deep Reinforcement Learning with Adversarial Attackscited by 152
2016EPOpt: Learning Robust Neural Network Policies Using Model Ensemblescited by 136
2021Robust Reinforcement Learning on State Observations with Learned Optimal Adversarycited by 133
2020Sponge Examples: Energy-Latency Attacks on Neural Networkscited by 108
2017Delving into adversarial attacks on deep policiescited by 104
2019Deep Reinforcement Learning for Cyber Securitycited by 92
2017Vulnerability of Deep Reinforcement Learning to Policy Induction Attackscited by 91
2020Adaptive Reward-Poisoning Attacks against Reinforcement Learningcited by 86
2020Policy Teaching via Environment Poisoning: Training-time Adversarial Attacks against Reinforcement Learningcited by 81
2018Reinforcement Learning with Perturbed Rewardscited by 76
2020Robust Deep Reinforcement Learning through Adversarial Losscited by 73
2019Policy Poisoning in Batch Reinforcement Learning and Controlcited by 70
2017Reinforcement Learning with a Corrupted Reward Channelcited by 68
2021BACKDOORL: Backdoor Attack against Competitive Reinforcement Learningcited by 66
2020Stealthy and Efficient Adversarial Attacks against Deep Reinforcement Learningcited by 62
2020Challenges and Countermeasures for Adversarial Attacks on Deep Reinforcement Learningcited by 61
2021Who Is the Strongest Enemy? Towards Optimal and Efficient Evasion Attacks in Deep RLcited by 54
2019Deceptive Reinforcement Learning Under Adversarial Manipulations on Cost Signalscited by 50
2021A Backdoor Attack against 3D Point Cloud Classifierscited by 50
2019Certified Adversarial Robustness for Deep Reinforcement Learningcited by 47
2021CROP: Certifying Robust Policies for Reinforcement Learning through Functional Smoothingcited by 41
2020Stop-and-Go: Exploring Backdoor Attacks on Deep Reinforcement Learning-based Traffic Congestion Control Systemscited by 35
2019Securing Connected & Autonomous Vehicles: Challenges Posed by Adversarial Machine Learning and The Way Forwardcited by 33
2019Characterizing Attacks on Deep Reinforcement Learningcited by 31
2020Robustifying Reinforcement Learning Agents via Action Space Adversarial Trainingcited by 30
2021Deep Learning-Based Autonomous Driving Systems: A Survey of Attacks and Defensescited by 29
2021Provably Efficient Black-Box Action Poisoning Attacks Against Reinforcement Learningcited by 29

50 most cited of 75 All 75 (JSON)

Built on by

Cited by year

Citing papers per year: 2024: 1.

1 paper.

2024Adaptive Discounting of Training Time Attackscited by 0

Similar, not cited

Close in wording, no citation either way