awesome-audit-algorithms

Audit algorithms

A curated collection of algorithms and research papers for auditing black-box machine learning models

A curated list of algorithms and papers for auditing black-box algorithms.

GitHub
94 stars
15 watching
9 forks
last commit: 3 months ago
Linked from 2 awesome lists

algorithmsauditawesomeawesome-listblackboxinferencereverse-engineering

Awesome Audit Algorithms / Papers / 2024
Auditing Local Explanations is Hard (NeurIPS)
LLMs hallucinate graphs too: a structural perspective (complex networks)
Fairness Auditing with Multi-Agent Collaboration (ECAI)
Mapping the Field of Algorithm Auditing: A Systematic Literature Review Identifying Research Trends, Linguistic and Geographical Disparities (Arxiv)
FairProof: Confidential and Certifiable Fairness for Neural Networks (Arxiv)
Under manipulations, are some AI models harder to audit? (SATML)
Improved Membership Inference Attacks Against Language Classification Models (ICLR)
Auditing Fairness by Betting (Neurips)

Awesome Audit Algorithms / Papers / 2023
Privacy Auditing with One (1) Training Run (NeurIPS - best paper)
Auditing fairness under unawareness through counterfactual reasoning (Information Processing & Management)
XAudit : A Theoretical Look at Auditing with Explanations (Arxiv)
Keeping Up with the Language Models: Robustness-Bias Interplay in NLI Data and Models (Arxiv)
Online Fairness Auditing through Iterative Refinement (KDD)
Stealing the Decoding Algorithms of Language Models (CCS)
Modeling rabbit‑holes on YouTube (SNAM)
Auditing YouTube’s Recommendation Algorithm for Misinformation Filter Bubbles (Transactions on Recommender Systems)
Auditing Yelp’s Business Ranking and Review Recommendation Through the Lens of Fairness (Arxiv)
Confidential-PROFITT: Confidential PROof of FaIr Training of Trees (ICLR)
SCALE-UP: An Efficient Black-box Input-level Backdoor Detection via Analyzing Scaled Prediction Consistency (ICLR)

Awesome Audit Algorithms / Papers / 2022
Two-Face: Adversarial Audit of Commercial Face Recognition Systems (ICWSM)
Scaling up search engine audits: Practical insights for algorithm auditing (Journal of Information Science)
A zest of lime: towards architecture-independent model distances (ICLR)
Active Fairness Auditing (ICML)
Look at the Variance! Efficient Black-box Explanations with Sobol-based Sensitivity Analysis (NeurIPS)
Your Echos are Heard: Tracking, Profiling, and Ad Targeting in the Amazon Smart Speaker Ecosystem (arxiv)

Awesome Audit Algorithms / Papers / 2021
When the Umpire is also a Player: Bias in Private Label Product Recommendations on E-commerce Marketplaces (FAccT)
Everyday Algorithm Auditing: Understanding the Power of Everyday Users in Surfacing Harmful Algorithmic Behaviors (CHI)
Auditing Black-Box Prediction Models for Data Minimization Compliance (NeurIPS)
Setting the Record Straighter on Shadow Banning (INFOCOM)
Extracting Training Data from Large Language Models (USENIX Security)
FairLens: Auditing black-box clinical decision support systems (Information Processing & Management)
Auditing Algorithmic Bias on Twitter (WebSci)
Bayesian Algorithm Execution: Estimating Computable Properties of Black-box Functions Using Mutual Information (ICML)

Awesome Audit Algorithms / Papers / 2020
Black-Box Ripper: Copying black-box models using generative evolutionary algorithms (NeurIPS)
Auditing radicalization pathways on (FAT*)
Adversarial Model Extraction on Graph Neural Networks (AAAI Workshop on Deep Learning on Graphs: Methodologies and Applications)
Remote Explainability faces the bouncer problem (Nature Machine Intelligence volume 2, pages529–539)
GeoDA: a geometric framework for black-box adversarial attacks (CVPR)
The Imitation Game: Algorithm Selectionby Exploiting Black-Box Recommender 2 2 months ago (Netys)
Auditing News Curation Systems:A Case Study Examining Algorithmic and Editorial Logic in Apple News (ICWSM)
Auditing Algorithms: On Lessons Learned and the Risks of DataMinimization (AIES)
Extracting Training Data from Large Language Models (arxiv)

Awesome Audit Algorithms / Papers / 2019
Adversarial Frontier Stitching for Remote Neural Network Watermarking (Neural Computing and Applications)
Knockoff Nets: Stealing Functionality of Black-Box Models (CVPR)
Opening Up the Black Box:Auditing Google's Top Stories Algorithm (Flairs-32)
Making targeted black-box evasion attacks effective andefficient (arXiv)
Online Learning for Measuring Incentive Compatibility in Ad Auctions (WWW)
TamperNN: Efficient Tampering Detection of Deployed Neural Nets (ISSRE)
Neural Network Model Extraction Attacks in Edge Devicesby Hearing Architectural Hints (arxiv)
Stealing Knowledge from Protected Deep Neural Networks Using Composite Unlabeled Data (ICNN)
Neural Network Inversion in Adversarial Setting via Background Knowledge Alignment (CCS)

Awesome Audit Algorithms / Papers / 2018
Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPR (Harvard Journal of Law & Technology)
Distill-and-Compare: Auditing Black-Box Models Using Transparent Model Distillation (AIES)
Towards Reverse-Engineering Black-Box Neural Networks (ICLR)
Data driven exploratory attacks on black box classifiers in adversarial domains (Neurocomputing)
xGEMs: Generating Examplars to Explain Black-Box Models (arXiv)
Learning Networks from Random Walk-Based Node Similarities (NIPS)
Identifying the Machine Learning Family from Black-Box Models (CAEPIA)
Stealing Neural Networks via Timing Side Channels (arXiv)
Copycat CNN: Stealing Knowledge by Persuading Confession with Random Non-Labeled Data (IJCNN)
Auditing the Personalization and Composition of Politically-Related Search Engine Results Pages (WWW)

Awesome Audit Algorithms / Papers / 2017
Uncovering Influence Cookbooks : Reverse Engineering the Topological Impact in Peer Ranking Services (CSCW)
The topological face of recommendation: models and application to bias detection (Complex Networks)
Membership Inference Attacks Against Machine Learning Models (Symposium on Security and Privacy)
Practical Black-Box Attacks against Machine Learning (Asia CCS)

Awesome Audit Algorithms / Papers / 2016
Algorithmic Transparency via Quantitative Input Influence: Theory and Experiments with Learning Systems (IEEE S&P)
Auditing Black-Box Models for Indirect Influence (ICDM)
Iterative Orthogonal Feature Projection for Diagnosing Bias in Black-Box Models (FATML Workshop)
Bias in Online Freelance Marketplaces: Evidence from TaskRabbit (dat workshop)
Stealing Machine Learning Models via Prediction APIs (Usenix Security)
“Why Should I Trust You?”Explaining the Predictions of Any Classifier (arXiv)
Back in Black: Towards Formal, Black Box Analysis of Sanitizers and Filters (Security and Privacy)
Algorithmic Transparency via Quantitative Input Influence: Theory and Experiments with Learning Systems (Security and Privacy)
An Empirical Analysis of Algorithmic Pricing on Amazon Marketplace (WWW)

Awesome Audit Algorithms / Papers / 2015
Certifying and Removing Disparate Impact (SIGKDD)
Peeking Beneath the Hood of Uber (IMC)

Awesome Audit Algorithms / Papers / 2014
A peek into the black box: exploring classifiers by randomization (Data Mining and Knowledge Discovery journal) ( )
XRay: Enhancing the Web's Transparency with Differential Correlation (USENIX Security)

Awesome Audit Algorithms / Papers / 2013
Measuring Personalization of Web Search (WWW)
Auditing: Active Learning with Outcome-Dependent Query Costs (NIPS)

Awesome Audit Algorithms / Papers / 2012
Query Strategies for Evading Convex-Inducing Classifiers (JMLR)

Awesome Audit Algorithms / Papers / 2008
Privacy Oracle: a System for Finding Application Leakswith Black Box Differential Testing (CCS)

Awesome Audit Algorithms / Papers / 2005
Adversarial Learning (KDD)
1st International Conference on Auditing and Artificial Intelligence
Regulatable ML Workshop (RegML'24)
Supporting User Engagement in Testing, Auditing, and Contesting AI (CSCW User AI Auditing)
Workshop on Algorithmic Audits of Algorithms (WAAA)
Regulatable ML Workshop (RegML'23)

Backlinks from these awesome lists:

0