MétaCan
Menu
Back to cohort
Record W4309447783 · doi:10.21203/rs.3.rs-2261000/v1

Membership Inference Attacks Against Temporally Correlated Data in Deep Reinforcement Learning

2022· preprint· en· W4309447783 on OpenAlex
Maziar Gomrokchi, Susan Amin, Hossein Aboutalebi, Alexander Wong, Doina Precup

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.
fundA Canadian funder is recorded on the work.

Bibliographic record

VenueResearch Square · 2022
Typepreprint
Languageen
FieldComputer Science
TopicAdversarial Robustness in Machine Learning
Canadian institutionsUniversity of WaterlooMcGill University
FundersNatural Sciences and Engineering Research Council of CanadaUniversity of WaterlooCompute Canada
KeywordsReinforcement learningAdversarial systemComputer scienceArtificial intelligenceInferenceReinforcementDeep learningMachine learningAdversaryVulnerability (computing)Temporal difference learningSet (abstract data type)Computer securityEngineering

Abstract

fetched live from OpenAlex

Abstract While significant research advances have been made in the field of deep reinforcement learning, there have been no concreteadversarial attack strategies in literature tailored for studying the vulnerability of deep reinforcement learning algorithms tomembership inference attacks. In such attacking systems, the adversary targets the set of collected input data on which thedeep reinforcement learning algorithm has been trained. To address this gap, we propose an adversarial attack frameworkdesigned for testing the vulnerability of a state-of-the-art deep reinforcement learning algorithm to a membership inferenceattack. In particular, we design a series of experiments to investigate the impact of temporal correlation, which naturally existsin reinforcement learning training data, on the probability of information leakage. Moreover, we compare the performance ofcollective and individual membership attacks against the deep reinforcement learning algorithm. Experimental results showthat the proposed adversarial attack framework is surprisingly effective at inferring data with an accuracy exceeding 84% inindividual and 97% in collective modes in three different continuous control Mujoco tasks, which raises serious privacy concernsin this regard. Finally, we show that the learning state of the reinforcement learning algorithm influences the level of privacybreaches significantly.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.011
metaresearch head score (Gemma)0.007
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesMeta-epidemiology (narrow), Open science, Research integrity
Consensus categoriesOpen science
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.965
Threshold uncertainty score1.000

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0110.007
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0010.000
Bibliometrics0.0010.002
Science and technology studies0.0010.000
Scholarly communication0.0010.001
Open science0.0120.057
Research integrity0.0010.014
Insufficient payload (model declined to judge)0.0010.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.138
GPT teacher head0.433
Teacher spread0.296 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it