Pith. sign in

REVIEW 5 cited by

Rethinking Adversarial Attacks in Reinforcement Learning from Policy Distribution Perspective

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2501.03562 v2 pith:2GAWCXIA submitted 2025-01-07 cs.LG cs.AI

Tianyang Duan , Zongyuan Zhang , Zheng Lin , Yue Gao , Ling Xiong , Yong Cui , Hongbin Liang , Xianhao Chen
show 2 more authors
Heming Cui Dong Huang
This is my paper · ORCID
classification cs.LGcs.AI
keywords policyattackdapgddistributionadversarialcomparedgradientindividual
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Deep Reinforcement Learning (DRL) suffers from uncertainties and inaccuracies in the observation signal in realworld applications. Adversarial attack is an effective method for evaluating the robustness of DRL agents. However, existing attack methods targeting individual sampled actions have limited impacts on the overall policy distribution, particularly in continuous action spaces. To address these limitations, we propose the Distribution-Aware Projected Gradient Descent attack (DAPGD). DAPGD uses distribution similarity as the gradient perturbation input to attack the policy network, which leverages the entire policy distribution rather than relying on individual samples. We utilize the Bhattacharyya distance in DAPGD to measure policy similarity, enabling sensitive detection of subtle but critical differences between probability distributions. Our experiment results demonstrate that DAPGD achieves SOTA results compared to the baselines in three robot navigation tasks, achieving an average 22.03% higher reward drop compared to the best baseline.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Defensive Adversarial CAPTCHA: A Semantics-Driven Framework for Natural Adversarial Example Generation

    cs.CV 2025-06 conditional novelty 6.0 of 10

    DAC and BP-DAC generate 'unsourced' adversarial CAPTCHAs from semantic prompts and report transfer attack success rates above 95% on ImageNet classifiers in black-box settings.

  2. DP-TRAE: A Dual-Phase Merging Transferable Reversible Adversarial Example for Image Privacy Protection

    cs.CR 2025-05 reject novelty 5.0 of 10

    DP-TRAE combines a transferable white-box adversarial initialization with a memory-augmented superpixel query attack and reversible embedding to mislead black-box image classifiers while allowing image recovery.

  3. Rethinking Membership Inference Attacks Against Transfer Learning

    cs.CR 2025-01 conditional novelty 5.0 of 10

    A white-box attack on the student model can infer teacher-training membership in transfer learning by comparing the student's hidden representations with those of a shadow student model.

  4. SFL-LEO: Asynchronous Split-Federated Learning Design for LEO Satellite-Ground Network Framework

    cs.NI 2025-04 reject novelty 4.0 of 10

    An asynchronous split-federated learning framework for LEO satellite-ground networks enables local updates during disconnection and heterogeneous model splitting, with simulated accuracy gains over split learning and ...

  5. Secure Resource Allocation via Constrained Deep Reinforcement Learning

    cs.LG 2025-01 reject novelty 3.0 of 10

    A deep Q-network with a fixed deadline penalty is claimed to cut simulated system cost by up to 40% and energy use by 41.5% in serverless multi-cloud offloading.

Pith tools