The Effect of Data Poisoning on Counterfactual Explanations | allainews.com

Feb. 14, 2024, 5:42 a.m. | Andr\'e Artelt Shubham Sharma Freddy Lecu\'e Barbara Hammer

cs.LG updates on arXiv.org arxiv.org

Counterfactual explanations provide a popular method for analyzing the predictions of black-box systems, and they can offer the opportunity for computational recourse by suggesting actionable changes on how to change the input to obtain a different (i.e. more favorable) system output. However, recent work highlighted their vulnerability to different types of manipulations. This work studies the vulnerability of counterfactual explanations to data poisoning. We formalize data poisoning in the context of counterfactual explanations for increasing the cost of recourse on …

box change computational counterfactual cs.ai cs.lg data data poisoning popular predictions systems types vulnerability work

More from arxiv.org / cs.LG updates on arXiv.org

Training towards significance with the decorrelated event classifier transformer neural network 1 day, 20 hours ago | arxiv.org

abstract analysis application arxiv +28

An adaptive standardisation methodology for Day-Ahead electricity price forecasting 1 day, 20 hours ago | arxiv.org

abstract algorithms arxiv complexity +18

SYNAuG: Exploiting Synthetic Data for Data Imbalance Problems 1 day, 20 hours ago | arxiv.org

abstract arxiv cs.cv cs.lg +17

Semantic Positive Pairs for Enhancing Visual Representation Learning of Instance Discrimination methods 1 day, 20 hours ago | arxiv.org

abstract algorithms arxiv augmentation +17

Description-Based Text Similarity 1 day, 20 hours ago | arxiv.org

abstract arxiv cases cs.cl +14

Improving Gradient Methods via Coordinate Transformations: Applications to Quantum Machine Learning 1 day, 20 hours ago | arxiv.org

abstract algorithms applications arxiv +13

A Generative Framework for Low-Cost Result Validation of Machine Learning-as-a-Service Inference 1 day, 20 hours ago | arxiv.org

abstract applications arxiv as-a-service +26

Digital Over-the-Air Federated Learning in Multi-Antenna Systems 1 day, 20 hours ago | arxiv.org

abstract arxiv communication computation +16

Bagging Provides Assumption-free Stability 1 day, 20 hours ago | arxiv.org

abstract algorithm arxiv assumptions +15

Data Architect

@ University of Texas at Austin | Austin, TX

View on ai-jobs.net

Data ETL Engineer

@ University of Texas at Austin | Austin, TX

View on ai-jobs.net

Lead GNSS Data Scientist

@ Lurra Systems | Melbourne

View on ai-jobs.net

Senior Machine Learning Engineer (MLOps)

@ Promaton | Remote, Europe

View on ai-jobs.net

Director, Clinical Data Science

@ Aura | Remote USA

View on ai-jobs.net

Research Scientist, AI (PhD)

@ Meta | Menlo Park, CA | New York City

View on ai-jobs.net