{"ID":2840743,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.13654","arxiv_id":"2511.13654","title":"Tuning for Two Adversaries: Enhancing the Robustness Against Transfer and Query-Based Attacks using Hyperparameter Tuning","abstract":"In this paper, we present the first detailed analysis of how training hyperparameters -- such as learning rate, weight decay, momentum, and batch size -- influence robustness against both transfer-based and query-based attacks. Supported by theory and experiments, our study spans a variety of practical deployment settings, including centralized training, ensemble learning, and distributed training. We uncover a striking dichotomy: for transfer-based attacks, decreasing the learning rate significantly enhances robustness by up to $64\\%$. In contrast, for query-based attacks, increasing the learning rate consistently leads to improved robustness by up to $28\\%$ across various settings and data distributions. Leveraging these findings, we explore -- for the first time -- the training hyperparameter space to jointly enhance robustness against both transfer-based and query-based attacks. Our results reveal that distributed models benefit the most from hyperparameter tuning, achieving a remarkable tradeoff by simultaneously mitigating both attack types more effectively than other training setups.","short_abstract":"In this paper, we present the first detailed analysis of how training hyperparameters -- such as learning rate, weight decay, momentum, and batch size -- influence robustness against both transfer-based and query-based attacks. Supported by theory and experiments, our study spans a variety of practical deployment setti...","url_abs":"https://arxiv.org/abs/2511.13654","url_pdf":"https://arxiv.org/pdf/2511.13654v2","authors":"[\"Pascal Zimmer\",\"Ghassan Karame\"]","published":"2025-11-17T18:03:49Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.CR\",\"cs.CV\"]","methods":"[]","has_code":false}