Towards both accurate and robust neural networks without extra data

Turnitin 降AI改写 早检测系统 早降重系统 Turnitin-UK版 万方检测-期刊版 维普编辑部版 Grammarly检测 Paperpass检测 checkpass检测 PaperYY检测
[Free GPT-4]
[DeepSeek]
[PDF] arxiv.org

Mixednuts: Training-free accuracy-robustness balance via nonlinearly mixed classifiers

Y Bai, M Zhou, VM Patel, S Sojoudi - arxiv preprint arxiv:2402.02263, 2024 - arxiv.org
Adversarial robustness often comes at the cost of degraded accuracy, impeding real-life
applications of robust classification models. Training-based solutions for better trade-offs are …
Spremi Citiraj Spominje se 8 puta Srodni članci Svih 4 inačica Prikaži kao HTML