提出学习动态的守恒与熵减定律，构建抗迁移攻击的终身集成学习方法。
Laws of Learning Dynamics and the Core of Learners
- 基于熵的终身集成学习，利用学习过程中的守恒与熵减规律
- 在CIFAR-10上对强扰动攻击下准确率提升显著，优于简单平均模型
- 适合防御对抗攻击的研究者，尤其关注模型鲁棒性与持续学习
我们提出了支配学习动态的基本规律，即守恒律与总熵减少律。在此框架下，引入一种基于熵的终身集成学习方法。通过构建免疫机制，评估其在抵御针对CIFAR-10数据集的迁移式对抗攻击中的有效性。相较于仅对干净样本和对抗样本分别训练后简单平均的朴素集成模型，所提出的logifold在多数测试场景中表现更优，尤其在强扰动条件下提升明显。
原文摘要 · Abstract (English)
We formulate the fundamental laws governing learning dynamics, namely the conservation law and the decrease of total entropy. Within this framework, we introduce an entropy-based lifelong ensemble learning method. We evaluate its effectiveness by constructing an immunization mechanism to defend against transfer-based adversarial attacks on the CIFAR-10 dataset. Compared with a naive ensemble formed by simply averaging models specialized on clean and adversarial samples, the resulting logifold achieves higher accuracy in most test cases, with particularly large gains under strong perturbations.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。