Deep generative models as an adversarial attack strategy for tabular machine learning

Salijona Dyrmishi, Mihaela Cătălina Stoian, Eleonora Giunchiglia, Maxime Cordy
9/19/2024
cs.LGcs.AI

No official code found. Best implementation: salijona/C-AdvDGM (1 star, MIT, last commit 2024-10-16).

Abstract

Deep Generative Models (DGMs) have found application in computer vision for generating adversarial examples to test the robustness of machine learning (ML) systems. Extending these adversarial techniques to tabular ML presents unique challenges due to the distinct nature of tabular data and the necessity to preserve domain constraints in adversarial examples. In this paper, we adapt four popular tabular DGMs into adversarial DGMs (AdvDGMs) and evaluate their effectiveness in generating realistic adversarial examples that conform to domain constraints.

View on arXivView PDF

Code Implementations(1)

salijona/C-AdvDGMCommunity100%
MIT
12Shell, PythonAug 7, 20242 years agoMIT

Similar Papers

TabularBench: Benchmarking Adversarial Robustness for Tabular Deep Learning in Real-world Use-cases

88% match1 repo232024

Constrained Adaptive Attacks: Realistic Evaluation of Adversarial Examples and Robust Training of Deep Neural Networks for Tabular Data

87% match2023

Insights on Adversarial Attacks for Tabular Machine Learning via a Systematic Literature Review

86% match2025

A Generative Adversarial Approach to Adversarial Attacks Guided by Contrastive Language-Image Pre-trained Model

83% match2025

CasTGAN: Cascaded Generative Adversarial Network for Realistic Tabular Data Synthesis

83% match1 repo92023

Adversarial Attacks Neutralization via Data Set Randomization

83% match2023

Cite this paper

@article{dyrmishi2024deep,
  title  = {Deep generative models as an adversarial attack strategy for tabular machine learning},
  author = {Salijona Dyrmishi and Mihaela Cătălina Stoian and Eleonora Giunchiglia and Maxime Cordy},
  year   = {2024},
  eprint = {2409.12642},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2409.12642v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2409.12642.svg)](https://codeofpaper.com/paper/2409.12642)

Discussion