Adversarial Attacks on Hyperbolic Networks
Not on the map. Why?
Abstract (arXiv)
As hyperbolic deep learning grows in popularity, so does the need for adversarial robustness in the context of such a non-Euclidean geometry. To this end, this paper proposes hyperbolic alternatives to the commonly used FGM and PGD adversarial attacks. Through interpretable synthetic benchmarks and experiments on existing datasets, we show how the existing and newly proposed attacks differ. Moreover, we investigate the differences in adversarial robustness between Euclidean and fully hyperbolic networks. We find that these networks suffer from different types of vulnerabilities and that the newly proposed hyperbolic attacks cannot address these differences. Therefore, we conclude that the shifts in adversarial robustness are due to the models learning distinct patterns resulting from their different geometries.
Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 61 of its 87 references that are in the map. First posted on arXiv 2 Dec 2024.
Builds on
61 of 87 references.
References · 61
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2013Intriguing properties of neural networkscited by 5,114
- 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
- 2020Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attackscited by 1,442
- 2016Geometric deep learning: going beyond Euclidean datacited by 1,042
- 2016Adversarial Machine Learning at Scalecited by 934
- 2020Fast is better than free: Revisiting adversarial trainingcited by 723
- 2019Square Attack: a query-efficient black-box adversarial attack via random searchcited by 682
- 2015Distillation as a Defense to Adversarial Perturbations against Deep Neural Networkscited by 675
- 2017Poincar'e Embeddings for Learning Hierarchical Representationscited by 674
38 more
- 2017Feature Squeezing: Detecting Adversarial Examples in Deep Neural Networkscited by 484
- 2019Hyperbolic Graph Convolutional Neural Networkscited by 400
- 2018Hyperbolic Neural Networkscited by 393
- 2018Feature Denoising for Improving Adversarial Robustnesscited by 312
- 2019Hyperbolic Graph Neural Networkscited by 244
- 2018Riemannian Adaptive Optimization Methodscited by 187
- 2018Hyperbolic Entailment Cones for Learning Hierarchical Embeddingscited by 160
- 2021Hyperbolic Deep Neural Networks: A Surveycited by 138
- 2023Hyperbolic Image-Text Representationscited by 132
- 2019Continuous Hierarchical Representations with Poincar'e Variational Auto-Encoderscited by 111
- 2022Hyperbolic Vision Transformers: Combining Improvements in Metric Learningcited by 109
- 2021Fully Hyperbolic Neural Networkscited by 107
- 2022Shadows can be Dangerous: Stealthy and Effective Physical-world Adversarial Attack by Natural Phenomenoncited by 101
- 2022Hyperbolic Image Segmentationcited by 101
- 2021Improving Adversarial Robustness via Channel-wise Activation Suppressingcited by 76
- 2023Hyperbolic Deep Learning in Computer Vision: A Surveycited by 72
- 2019Hyperbolic Graph Attention Networkcited by 69
- 2021Lorentzian Graph Convolutional Networkscited by 66
- 2021FIBA: Frequency-Injection based Backdoor Attack in Medical Image Analysiscited by 60
- 2021Clipped Hyperbolic Classifiers Are Super-Hyperbolic Classifierscited by 56
- 2022Label-Only Model Inversion Attacks via Boundary Repulsioncited by 56
- 2021A Hyperbolic-to-Hyperbolic Graph Convolutional Networkcited by 53
- 2020Latent Variable Modelling with Hyperbolic Normalizing Flowscited by 49
- 2021Learning the Predictability of the Futurecited by 48
- 2022Towards Robust Rain Removal Against Adversarial Attacks: A Comprehensive Benchmark Analysis and Beyondcited by 48
- 2023Fully Hyperbolic Convolutional Neural Networks for Computer Visioncited by 48
- 2020Hierarchical Image Classification using Entailment Cone Embeddingscited by 46
- 2023HYperbolic Self-Paced Learning for Self-Supervised Skeleton-based Action Representationscited by 42
- 2018Large-Margin Classification in Hyperbolic Spacecited by 41
- 2023Physically Realizable Natural-Looking Clothing Textures Evade Person Detectors via 3D Modelingcited by 41
- 2021Hyperbolic Busemann Learning with Ideal Prototypescited by 40
- 2021Unsupervised Discovery of the Long-Tail in Instance Segmentation Using Hierarchical Self-Supervisioncited by 39
- 2021Dual-Key Multimodal Backdoors for Visual Question Answeringcited by 35
- 2023Towards Effective Adversarial Textured 3D Meshes on Physical Face Recognitioncited by 33
- 2020HyperText: Endowing FastText with Hyperbolic Geometrycited by 28
- 2022Hyperbolic Deep Reinforcement Learningcited by 25
- 2020Capturing implicit hierarchical structure in 3D biomedical images with self-supervised hyperbolic representationscited by 24
- 2023Poincar'e ResNetcited by 24
50 most cited of 61 All 61 (JSON)
Built on by
Citing papers per year: 2025: 1; 2026: 2.
3 papers.