Adaptive Epsilon Adversarial Training for Robust Gravitational Wave Parameter Estimation Using Normalizing Flows
This work addresses robustness in gravitational wave analysis, an important but relatively narrow domain, with incremental improvements to existing adversarial training techniques.
The paper tackles the problem of improving robustness of Normalizing Flow models for gravitational wave parameter estimation against adversarial attacks by proposing an adaptive epsilon method for FGSM adversarial training that dynamically adjusts perturbation strengths based on gradient magnitudes. Their hybrid ResNet and Inverse Autoregressive Flow architecture reduces Negative Log Likelihood loss by 47% under FGSM attacks compared to baseline while maintaining competitive performance on clean data.
Adversarial training with Normalizing Flow (NF) models is an emerging research area aimed at improving model robustness through adversarial samples. In this study, we focus on applying adversarial training to NF models for gravitational wave parameter estimation. We propose an adaptive epsilon method for Fast Gradient Sign Method (FGSM) adversarial training, which dynamically adjusts perturbation strengths based on gradient magnitudes using logarithmic scaling. Our hybrid architecture, combining ResNet and Inverse Autoregressive Flow, reduces the Negative Log Likelihood (NLL) loss by 47\% under FGSM attacks compared to the baseline model, while maintaining an NLL of 4.2 on clean data (only 5\% higher than the baseline). For perturbation strengths between 0.01 and 0.1, our model achieves an average NLL of 5.8, outperforming both fixed-epsilon (NLL: 6.7) and progressive-epsilon (NLL: 7.2) methods. Under stronger Projected Gradient Descent attacks with perturbation strength of 0.05, our model maintains an NLL of 6.4, demonstrating superior robustness while avoiding catastrophic overfitting.