提出无需调参的自适应归一化方法，提升长尾识别性能
Why Not Hyperparameter-Friendly Optimisation? A Monotonic Adaptive Norm Rescaling Approach For Long-Tailed Recognition
- 用单调性约束直接调整类别权重，避免传统正则化调参
- 在多个基准数据集上显著提升长尾识别准确率，达最优水平
- 方法通用易集成，适合各类长尾学习场景
长尾识别是深度学习中的重大挑战。两阶段解耦范式（先表征学习，再分类器重训练）提供有效解决方案。分类器重训练阶段常用自适应范数重缩放技术，通过参数正则化调整每类权重范数，但引入了依赖敏感的超参数。本文从类别条件分布角度分析支持该方法，并提出简单有效的自适应单调归一化（SAMN）方法。SAMN 不依赖参数正则化，而是利用池相邻违规者算法直接强制每类权重范数单调递增，实现超参数友好。该方法为通用策略，可无缝融合其他方法以进一步提升性能。在多个基准数据集上的实验表明，该方法显著改善长尾识别表现，常达到当前最优结果。
原文摘要 · Abstract (English)
Long-tailed recognition poses a significant challenge for deep learning. The two-stage decoupling paradigm, which separates representation learning from classifier retraining, offers a promising solution. During the classifier retraining stage, adaptive norm rescaling is a popular technique. It adjusts the per-class weight norms via parameter regularization, which inevitably introduces hyperparameters. However, many studies report that long-tailed recognition is sensitive to these hyperparameters, as their setup significantly impacts performance. In this paper, we first provide a class-conditional distribution perspective to support norm rescaling methods. Furthermore, we propose a simple but effective approach called Self-Adaptive Monotonic Normalization (SAMN). SAMN avoids the need for parameter regularization. It directly enforces monotonicity on per-class weight norms using the Pool Adjacent Violators Algorithm, making the method hyperparameter-friendly. SAMN is a universal strategy that integrates seamlessly with other methods for enhanced performance. Experiments on benchmark datasets demonstrate that our method significantly boosts long-tailed recognition performance, often achieving state-of-the-art results.
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。