优化图像分类数据分布，让模型更关注关键类别。
Optimizing Class Distributions for Bias-Aware Multi-Class Learning
- 通过迭代方法寻找多类任务中最优的样本数量分配。
- 在CIFAR-10和iNaturalist21上提升模型整体与平衡性能。
- 适合对特定类别可靠性要求高的安全场景使用。
我们提出BiCDO（Bias-Controlled Class Distribution Optimizer），一种迭代式、以数据为中心的框架，用于为多分类图像识别任务寻找帕累托最优的类别分布。该方法可对特定类别进行性能优先级设置，适用于安全关键场景（如优先识别‘人类’而非‘狗’）。相较于均匀分布，BiCDO能确定每类最优图像数量，从而提升模型可靠性并降低目标函数中的偏差与方差。该方法可无缝集成至现有训练流程，仅需少量代码修改，并支持任意标注的多类数据集。我们在CIFAR-10和iNaturalist21上使用EfficientNet、ResNet和ConvNeXt验证了BiCDO，结果表明通过优化数据分布可实现更优且更均衡的模型表现。
原文摘要 · Abstract (English)
We propose BiCDO (Bias-Controlled Class Distribution Optimizer), an iterative, data-centric framework that identifies Pareto optimized class distributions for multi-class image classification. BiCDO enables performance prioritization for specific classes, which is useful in safety-critical scenarios (e.g. prioritizing 'Human' over 'Dog'). Unlike uniform distributions, BiCDO determines the optimal number of images per class to enhance reliability and minimize bias and variance in the objective function. BiCDO can be incorporated into existing training pipelines with minimal code changes and supports any labelled multi-class dataset. We have validated BiCDO using EfficientNet, ResNet and ConvNeXt on CIFAR-10 and iNaturalist21 datasets, demonstrating improved, balanced model performance through optimized data distribution.
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。