arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2309.02476stat.MLcs.LG
收藏

基于不确定性估计的最优样本选择及其在深度学习中的应用

Optimal Sample Selection Through Uncertainty Estimation and Its Application in Deep Learning

The Hong Kong University of Science and Technology(香港科技大学)

Yong Lin, Chen Liu, Chenlu Ye, Qing Lian, Yuan Yao, Tong Zhang

AI总结:

针对深度学习数据标注成本高的问题,提出基于不确定性的最优子采样方法COPS,利用模型logits估计采样比率,同时解决核心集选择与主动学习,实验证明其优于基线方法。

摘要:

现代深度学习严重依赖大型标注数据集,这通常伴随着人工标注和计算资源方面的高昂成本。为缓解这些挑战,研究人员探索了信息性子集选择技术,包括核心集选择和主动学习。具体而言,核心集选择涉及对同时包含输入($\x$)和输出($\y$)的数据进行采样,而主动学习则仅关注输入数据($\x$)。\n在本研究中,我们在线性softmax回归的背景下,提出了一个理论上最优的解决方案,以同时解决核心集选择和主动学习问题。我们提出的方法COPS(基于不确定性的最优子采样)旨在最小化在子采样数据上训练的模型的期望损失。与依赖逆协方差矩阵显式计算的现有方法不同(这些方法不易适用于深度学习场景),COPS利用模型的logits来估计采样比率。该采样比率与模型不确定性密切相关,并能有效应用于深度学习任务。此外,我们通过引入对低密度样本的降权方法来解决模型对错误设定的敏感性问题,这一思路借鉴了先前的工作。\n为评估所提方法的有效性,我们在基准数据集上使用深度神经网络进行了大量实证实验。结果一致表明,COPS相比基线方法具有更优越的性能,再次证实了其有效性。

英文摘要:

Modern deep learning heavily relies on large labeled datasets, which often comse with high costs in terms of both manual labeling and computational resources. To mitigate these challenges, researchers have explored the use of informative subset selection techniques, including coreset selection and active learning. Specifically, coreset selection involves sampling data with both input ($\bx$) and output ($\by$), active learning focuses solely on the input data ($\bx$). In this study, we present a theoretically optimal solution for addressing both coreset selection and active learning within the context of linear softmax regression. Our proposed method, COPS (unCertainty based OPtimal Sub-sampling), is designed to minimize the expected loss of a model trained on subsampled data. Unlike existing approaches that rely on explicit calculations of the inverse covariance matrix, which are not easily applicable to deep learning scenarios, COPS leverages the model's logits to estimate the sampling ratio. This sampling ratio is closely associated with model uncertainty and can be effectively applied to deep learning tasks. Furthermore, we address the challenge of model sensitivity to misspecification by incorporating a down-weighting approach for low-density samples, drawing inspiration from previous works. To assess the effectiveness of our proposed method, we conducted extensive empirical experiments using deep neural networks on benchmark datasets. The results consistently showcase the superior performance of COPS compared to baseline methods, reaffirming its efficacy.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑