考虑主动学习中样本依赖性，提升模型参数估计精度
Dependency-aware Maximum Likelihood Estimation for Active Learning
- 提出DMLE方法，修正传统MLE忽略样本依赖的问题
- 在前100个样本内，准确率提升6%~10.5%（批量查询量k=1~10）
- 适合关注主动学习优化与模型训练一致性的研究者
主动学习通过策略性地选择样本以高效构建标注数据集。在此序列化过程中，每个样本的获取会影响后续选择，导致标注集中的样本存在依赖关系。然而，在使用最大似然估计（MLE）更新模型时，传统方法假设数据独立同分布（i.i.d.），忽略了这种依赖性。本文提出依赖感知的最大似然估计（DMLE），在主动学习框架内修正MLE，解决因i.i.d.假设被忽略而产生的样本依赖问题，使模型参数估计更符合主动学习原则。实验显示，该方法在多个基准数据集上表现更优，在收集前100个样本后，采用熵作为选择函数，批量查询大小k=1、5、10时，平均准确率分别提升6%、8.6%和10.5%。代码已公开于：https://github.com/neu-spiral/DMLEforAL
原文摘要 · Abstract (English)
Active learning aims to efficiently build a labeled training set by strategically selecting samples to query labels from annotators. In this sequential process, each sample acquisition influences subsequent selections, causing dependencies among samples in the labeled set. However, these dependencies are overlooked during the model parameter estimation stage when updating the model using Maximum Likelihood Estimation (MLE), a conventional method that assumes independent and identically distributed (i.i.d.) data. We propose Dependency-aware MLE (DMLE), which corrects MLE within the active learning framework by addressing sample dependencies typically neglected due to the i.i.d. assumption, ensuring consistency with active learning principles in the model parameter estimation process. This improved method achieves superior performance across multiple benchmark datasets, reaching higher performance in earlier cycles compared to conventional MLE. Specifically, we observe average accuracy improvements of 6%, 8.6%, and 10.5% for k=1, k=5, and k=10 respectively, after collecting the first 100 samples, where entropy is the acquisition function and k is the query batch size acquired at every active learning cycle. Our implementation is publicly available at: https://github.com/neu-spiral/DMLEforAL
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。