Research map / Adversarial robustness and security / Attacks on text and graphs
Adversarial Attack for Asynchronous Event-Based Data
Wooju Lee, Hyun Myung
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add generation of adversarial events to event-time shifting (ablation) | Event attack · N-Caltech101, untargeted, most tested representations | attack success rate (%): 95.50 -> 95.87; 96.13 -> 100.0; 95.96 -> 99.0; 95.14 -> 96.12; 94.16 -> 96.03 | Improved |
| add generation of adversarial events to event-time shifting (ablation) | Event attack · N-Caltech101, random targeted, EST (10), EST (5), Voxel grid (10), MLP and EST (10), Trilinear | attack success rate (%): 41.64 -> 69.50; 63.87 -> 70.74; 52.62 -> 80.23; 41.42 -> 69.78 | Improved |
| add generation of adversarial events to shifting (ablation) | Event attack · N-Caltech101, random targeted, Two-channel (1) | attack success rate (%): 69.87 -> 69.48 | Hurt |
| add generation of adversarial events to shifting (ablation) | Event attack · N-Caltech101, random targeted, Two-channel (1), Trilinear | attack success rate (%): 67.84 -> 67.42 | Hurt |
| use adversarial training on original and adversarial events | EST (5) ResNet-34 · N-Caltech101, shifting attack | top-1 accuracy (%): 2.99 -> 52.79 | Improved |
| use adversarial training on original and adversarial events | EST (5) ResNet-34 · N-Caltech101, shifting and generating attack | top-1 accuracy (%): 0 -> 22.86 | Improved |
| use adversarial training rather than original-only training | EST (5) ResNet-34 · N-Caltech101, clean | top-1 accuracy (%): 84.99 -> 80.67 | Hurt |
| increase the shifting perturbation size (ablation) | EST (10) · N-Caltech101, untargeted attack | attack success rate (%): 95.11; 95.50; 93.40; 94.96; 90.66 | Mixed |
| increase the shifting perturbation size (ablation) | EST (1) · N-Caltech101, untargeted attack | attack success rate (%): 62.61; 86.01; 93.63; 95.96; 98.05 | Improved |
| reduce event-data frequency by half (ablation) | EST (10) shifting attack · N-Caltech101, untargeted attack, epsilon 0.05 | attack success rate (%): 95.11 -> 75.5 | Hurt |
| use adversarial events from EST (10) or Voxel grid (10) against EST (1) | Cross-representation attack · N-Caltech101, untargeted transfer | attack success rate (%): 0.72; 0.86 | Hurt |
| use adversarial events from EST (1) against other representations | Cross-representation attack · N-Caltech101, untargeted transfer | attack success rate (%): 27.51; 46.85 | Improved |
Closest papers by meaning
- Exploring Vulnerabilities in Spiking Neural Networks: Direct Adversarial Attacks on Raw Event Data (ECCV 2024 · similarity 0.87)
- Robust Adversarial Objects against Deep Learning Models (AAAI 2020 · similarity 0.87)
- Minimal Adversarial Examples for Deep Learning on 3D Point Clouds (ICCV 2021 · similarity 0.85)
- Generating 3D Adversarial Point Clouds (CVPR 2019 · similarity 0.85)
- Differentiable Adversarial Attacks for Marked Temporal Point Processes (AAAI 2025 · similarity 0.84)
- Seeing is Not Believing: Adversarial Natural Object Optimization for Hard-Label 3D Scene Attacks (CVPR 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
