paper-with-me

Papers

Learning to Deceive Knowledge Graph Augmented Models via Targeted Perturbation

2020-10-24 · ICLR 2021 1 · Mrigank Raman, Aaron Chan, Siddhant Agarwal, Peifeng Wang, Hansen Wang, Sungchul Kim, Ryan Rossi, Handong Zhao, Nedim Lipka, Xiang Ren

Knowledge graphs (KGs) have helped neural models improve performance on various knowledge-intensive tasks, like question answering and item recommendation. By using attention over the KG, such KG-augmented models can also "explain" which KG information was most relevant for making a given prediction. In this paper, we question whether these models are really behaving as we expect. We show that, through a reinforcement learning policy (or even simple heuristics), one can produce deceptively perturbed KGs, which maintain the downstream performance of the original KG while significantly deviating from the original KG's semantics and structure. Our findings raise doubts about KG-augmented models' ability to reason about KG information and give sensible explanations.

📄 PDF Abstract BibTeX arXiv:2010.12872

Code (1)

INK-USC/deceive-KG-models 공식 구현 pytorch

Tasks

Knowledge GraphsQuestion AnsweringRecommendation Systemsreinforcement-learningReinforcement Learning (RL)

Similar Papers 제목 키워드 기반

Indirect Adversarial Attacks via Poisoning Neighbors for Graph Convolutional Networks

2020-02-19 · Tsubasa Takahashi

Graph convolutional neural networks, which learn aggregations over neighbor nodes, have achieved great performance in node classification tasks. However, recent studies reported that such graph convolutional node classif…

General ClassificationNode Classification

Single-Class Target-Specific Attack against Interpretable Deep Learning Systems

2023-07-12 · Eldor Abdukhamidov, Mohammed Abuhamad, George K. Thiruvathukal, Hyoungshick Kim 외

In this paper, we present a novel Single-class target-specific Adversarial attack called SingleADV. The goal of SingleADV is to generate a universal perturbation that deceives the target model into confusing a specific c…

Adversarial AttackDeep Learning

Introducing Competition to Boost the Transferability of Targeted Adversarial Examples through Clean Feature Mixup

2023-05-24 · CVPR 2023 1 · Junyoung Byun, Myung-Joon Kwon, Seungju Cho, Yoonji Kim 외

Deep neural networks are widely known to be susceptible to adversarial examples, which can cause incorrect predictions through subtle input modifications. These adversarial examples tend to be transferable between models…

Rethinking Impersonation and Dodging Attacks on Face Recognition Systems

2024-01-17 · Fengfan Zhou, Qianyu Zhou, Bangjie Yin, Hui Zheng 외

Face Recognition (FR) systems can be easily deceived by adversarial examples that manipulate benign face images through imperceptible perturbations. Adversarial attacks on FR encompass two types: impersonation (targeted)…

Adversarial AttackFace Recognition

Decision-BADGE: Decision-based Adversarial Batch Attack with Directional Gradient Estimation

2023-03-09 · Geunhyeok Yu, Minwoo Jeon, Hyoseok Hwang

The susceptibility of deep neural networks (DNNs) to adversarial examples has prompted an increase in the deployment of adversarial attacks. Image-agnostic universal adversarial perturbations (UAPs) are much more threate…

Adversarial Attack