针对长尾数据分类性能下降问题，提出一种基于贝叶斯理论的简单修正方法。
Class-Balanced Softmax: A Bayes Theory-Based Method for Long-Tailed Recognition
- 基于贝叶斯框架和幂律假设，对logit进行轻量级调整。
- 在多个大规模数据集上显著提升尾部类别准确率，优于现有方法。
- 适用于长尾识别场景，尤其适合追求高精度尾类分类的研究者。
使用传统softmax分类器的深度学习模型在各类分类任务中取得了显著成功，但在数据不平衡情况下性能大幅下降。尽管平衡softmax被广泛采用为最先进的重平衡方法，但仍存在内在局限，例如导致尾部类别测试准确率过低。为此，我们提出类平衡softmax（CBS）。CBS基于理论贝叶斯框架和启发式幂律假设，是一种计算开销小、可轻松集成到现有流程中的简单logit调整方法。此外，我们揭示了在不平衡数据上训练模型时的一个基本现象——偏好问题，即数据量少的类别具有更高的训练误差和更大的泛化差距。为量化该问题，我们引入新指标，并证明CBS能有效缓解偏好问题。在大规模基准上的大量实验表明，CBS具有高度可扩展性，性能优于现有方法，包括平衡softmax。
原文摘要 · Abstract (English)
Deep learning models using traditional softmax classifiers have achieved remarkable success in various classification tasks. However, their performance degrades significantly on imbalanced datasets. Although Balanced Softmax is widely adopted as a state-of-the-art rebalancing method, it possesses inherent limitations, such as yielding disproportionately lower testing accuracy for tail classes. To mitigate these shortcomings, we propose the Class-Balanced Softmax (CBS). Rooted in a theoretical Bayesian framework and a heuristic power-law assumption, the CBS is a simple logit adjustment that is computationally inexpensive and easily integrated into existing pipelines. Furthermore, we characterise a fundamental phenomenon in models trained on imbalanced data, termed the preference issue, wherein models exhibit higher training error and a larger generalisation gap for classes with limited data. To quantify this issue, we introduce a novel metric and demonstrate that CBS effectively mitigates the preference issue. Extensive experiments on large-scale benchmarks show that CBS is highly scalable and outperforms existing methods, including Balanced Softmax.
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。