arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2006.04381cs.LGstat.MEstat.ML
收藏

平衡子采样稳定预测

Balance-Subsampled Stable Prediction

Zhejiang University(浙江大学);The University of Hong Kong(香港大学)

Kun Kuang, Hengtao Zhang, Fei Wu, Yueting Zhuang, Aijun Zhang

摘要:

在机器学习中,通常假设训练数据和测试数据共享相同的总体分布。然而,在实际应用中,由于样本选择偏差会导致训练数据向测试数据的分布偏移,这一假设常常被违背。这种与模型无关的分布偏移通常会导致未知测试数据的预测不稳定。本文基于部分因子设计理论,提出了一种新的平衡下采样稳定预测(BSSP)算法。它将每个预测因子的明确效应与混杂变量隔离开来。设计理论分析表明,该方法能有效地降低由分布偏移引起的预测变量间的混杂效应,从而提高参数估计的精度和预测的稳定性。在合成数据集和真实数据集上的数值实验表明,我们的BSSP算法在未知测试数据上的稳定预测性能明显优于基线方法。

英文摘要:

In machine learning, it is commonly assumed that training and test data share the same population distribution. However, this assumption is often violated in practice because the sample selection bias may induce the distribution shift from training data to test data. Such a model-agnostic distribution shift usually leads to prediction instability across unknown test data. In this paper, we propose a novel balance-subsampled stable prediction (BSSP) algorithm based on the theory of fractional factorial design. It isolates the clear effect of each predictor from the confounding variables. A design-theoretic analysis shows that the proposed method can reduce the confounding effects among predictors induced by the distribution shift, hence improve both the accuracy of parameter estimation and prediction stability. Numerical experiments on both synthetic and real-world data sets demonstrate that our BSSP algorithm significantly outperforms the baseline methods for stable prediction across unknown test data.

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。

↑