ASAT: Adaptively Scaled Adversarial Training in Time Series
This work addresses the need for more robust and generalizable neural networks in time series analysis, particularly for financial applications, but it is incremental as it builds on existing adversarial training methods.
The authors tackled the problem of improving neural network generalization and adversarial robustness in time series analysis, specifically in finance, by proposing Adaptively Scaled Adversarial Training (ASAT), which rescales data at different time slots with adaptive scales. Experimental results show that ASAT improves generalization ability and adversarial robustness compared to baselines, achieving better generalization than traditional adversarial training with similar robustness.
Adversarial training is a method for enhancing neural networks to improve the robustness against adversarial examples. Besides the security concerns of potential adversarial examples, adversarial training can also improve the generalization ability of neural networks, train robust neural networks, and provide interpretability for neural networks. In this work, we introduce adversarial training in time series analysis to enhance the neural networks for better generalization ability by taking the finance field as an example. Rethinking existing research on adversarial training, we propose the adaptively scaled adversarial training (ASAT) in time series analysis, by rescaling data at different time slots with adaptive scales. Experimental results show that the proposed ASAT can improve both the generalization ability and the adversarial robustness of neural networks compared to the baselines. Compared to the traditional adversarial training algorithm, ASAT can achieve better generalization ability and similar adversarial robustness.