提出FIRAL算法，用信息率优化提升多分类逻辑回归的主动学习性能。
FIRAL: An Active Learning Algorithm for Multinomial Logistic Regression
- 基于有限样本分析，用费舍尔信息率控制分类误差上界。
- 在MNIST、CIFAR-10等数据集上，分类错误率低于其他五种方法。
- 适合需要高精度少标注训练的多分类主动学习场景。
我们研究了基于池的主动学习在多分类任务中使用多项式逻辑回归的理论与算法。通过有限样本分析，证明费舍尔信息率（FIR）可上下界界定超额风险。基于该理论，提出一种利用遗憾最小化来最小化FIR的主动学习算法——FIRAL。为验证推导的超额风险边界，我们在合成数据集上进行了实验。此外，将FIRAL与五种其他方法对比，结果表明：在多分类逻辑回归设置下，其始终产生最小的分类误差，这一结论在MNIST、CIFAR-10和50类ImageNet数据集上的实验中得到证实。
原文摘要 · Abstract (English)
We investigate theory and algorithms for pool-based active learning for multiclass classification using multinomial logistic regression. Using finite sample analysis, we prove that the Fisher Information Ratio (FIR) lower and upper bounds the excess risk. Based on our theoretical analysis, we propose an active learning algorithm that employs regret minimization to minimize the FIR. To verify our derived excess risk bounds, we conduct experiments on synthetic datasets. Furthermore, we compare FIRAL with five other methods and found that our scheme outperforms them: it consistently produces the smallest classification error in the multiclass logistic regression setting, as demonstrated through experiments on MNIST, CIFAR-10, and 50-class ImageNet.
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。