Adversarial Training Improves Generalization Under Distribution Shifts In Bioacoustics

Abstract

Adversarial training is a promising strategy for enhancing model robustness against adversarial attacks. However, its impact on generalization under substantial data distribution shifts in audio classification remains largely unexplored. To address this gap, this work investigates how different adversarial training strategies improve generalization performance and adversarial robustness in audio classification. The study focuses on two model architectures: a conventional convolutional neural network (ConvNeXt) and an inherently interpretable prototype-based model (AudioProtoPNet). The approach is evaluated using a challenging bird sound classification benchmark. This benchmark is characterized by pronounced distribution shifts between training and test data due to varying environmental conditions and recording methods, a common real-world challenge. The investigation explores two adversarial training strategies: one based on output-space attacks that maximize the classification loss fu

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyheinrich2025adversarial

Related papers

• Boosting Noise Robustness Of Acoustic Model Via Deep Adversarial Training (2018)9.23
• Targeted Adversarial Examples For Black Box Audio Systems (2018)15.75
• Adversarial Learning Of Raw Speech Features For Domain Invariant Speech Recognition (2018)9.23
• Training Augmentation With Adversarial Examples For Robust Speech Recognition (2018)11.39
• Attentive Adversarial Learning For Domain-invariant Training (2019)6.77
• Adversarial Attack And Defense Strategies For Deep Speaker Recognition Systems (2020)13.39
• Inaudible Adversarial Perturbations For Targeted Attack In Speaker Recognition (2020)12.33
• Towards Understanding And Mitigating Audio Adversarial Examples For Speaker Recognition (2022)11.67