Research map / Time series / Causal and generative time series
Adversarial Attacks on Time Series
Fazle Karim, Somshubra Majumdar, Houshang Darabi
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 |
|---|---|---|---|
| change target class from 1 to 2 (ablation) | GATN black-box attack on 1-NN DTW · UCR Trace | number of adversaries: 0 -> 9 | Improved |
| change target class from 1 to 2 (ablation) | GATN white-box attack on 1-NN DTW · UCR Trace | number of adversaries: 0 -> 3 | Improved |
| change target class from 1 to 2 (ablation) | GATN black-box attack on FCN · UCR Trace | number of adversaries: 0 -> 1 | Improved |
| change target class from 1 to 2 (ablation) | GATN white-box attack on FCN · UCR Trace | number of adversaries: 1 -> 37 | Improved |
| use white-box instead of black-box access | GATN on FCN · UCR Wafer | number of adversaries: 543 -> 2805 | Improved |
| use white-box instead of black-box access | GATN on FCN · UCR StarLightCurves | number of adversaries: 577 -> 899 | Improved |
| use white-box instead of black-box access | GATN on 1-NN DTW · UCR Wafer | number of adversaries: 1843 -> 1295 | Hurt |
| use white-box instead of black-box access | GATN on 1-NN DTW · UCR FordB | number of adversaries: 238 -> 223 | No clear effect |
| attack FCN instead of 1-NN DTW (black-box) | GATN · UCR 42 datasets | number of adversaries: p=6.706E-04 | Hurt |
| attack FCN white-box instead of black-box | GATN · UCR 42 datasets | number of adversaries: p=5.680E-07 | Improved |
| use white-box instead of black-box access | GATN on FCN · UCR 42 datasets | MSE: p=5.694E-01 | No clear effect |
| use black-box 1-NN DTW versus black-box FCN | GATN · UCR 42 datasets | MSE: p=1.843E-02 | Mixed |
Closest papers by meaning
- Dynamic Time Warping Based Adversarial Framework for Time-Series Domain (TPAMI 2022 · similarity 0.90)
- Black-Box Adversarial Attack on Time Series Classification (AAAI 2023 · similarity 0.88)
- Fortifying Time Series: DTW-Certified Robust Anomaly Detection (NeurIPS 2025 · similarity 0.86)
- TVNet: A Novel Time Series Analysis Method Based on Dynamic Convolution and 3D-Variation (ICLR 2025 · similarity 0.84)
- Spear and Shield: Adversarial Attacks and Defense Methods for Model-Based Link Prediction on Continuous-Time Dynamic Graphs (AAAI 2024 · similarity 0.83)
- BackTime: Backdoor Attacks on Multivariate Time Series Forecasting (NeurIPS 2024 · 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.
