LGAICRJan 9, 2023

On the Susceptibility and Robustness of Time Series Models through Adversarial Attack and Defense

arXiv:2301.03703v17 citationsh-index: 5
Originality Synthesis-oriented
AI Analysis

This work addresses the robustness of time series models for applications like forecasting or anomaly detection, but it is incremental as it applies existing adversarial methods to time series data.

The study evaluated the vulnerability of seven time series models to three adversarial attacks and their recovery using one defense, finding that all models, especially GRU and RNN, are susceptible, with FGSM being the most effective attack and PGD the hardest to recover from.

Under adversarial attacks, time series regression and classification are vulnerable. Adversarial defense, on the other hand, can make the models more resilient. It is important to evaluate how vulnerable different time series models are to attacks and how well they recover using defense. The sensitivity to various attacks and the robustness using the defense of several time series models are investigated in this study. Experiments are run on seven-time series models with three adversarial attacks and one adversarial defense. According to the findings, all models, particularly GRU and RNN, appear to be vulnerable. LSTM and GRU also have better defense recovery. FGSM exceeds the competitors in terms of attacks. PGD attacks are more difficult to recover from than other sorts of attacks.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes