PrIU: A Provenance-Based Approach for Incrementally Updating Regression Models

PrIU: A Provenance-Based Approach for Incrementally Updating Regression Models
复制标题

PrIU:一种基于来源的增量更新回归模型的方法

DOI:
10.1145/3318464.3380571
复制
发表时间:
2020
期刊:
ISBN 978-1-4503-6735-6
影响因子:
--
通讯作者:
Davidson, Susan
Davidson, Susan
中科院分区:
--
文献类型:
--
作者:
Wu, Yinjun;Tannen, Val;Davidson, Susan

文献摘要

参考文献

被引文献

相似文献

机器学习算法的广泛应用给传统的数据库问题如增量视图更新带来了新的挑战。人们正在努力更好地理解和调试机器学习模型,以及识别和修复训练数据集中的错误。我们的重点是如何帮助这些活动,当他们在清理或选择不同的训练数据子集以实现可解释性时,在删除有问题的训练样本后,必须重新训练机器学习模型。本文提出了一种有效的基于种源的方法,PrIU,其优化版本,PrIU-opt,增量更新模型参数,而不牺牲预测精度。我们证明了增量更新模型参数的正确性和收敛性,并通过实验验证。实验结果表明,与简单地从头开始重新训练模型相比,PrIU-opt可以实现高达两个数量级的加速,但获得高度相似的模型。
The ubiquitous use of machine learning algorithms brings new challenges to traditional database problems such as incremental view update. Much effort is being put in better understanding and debugging machine learning models, as well as in identifying and repairing errors in training datasets. Our focus is on how to assist these activities when they have to retrain the machine learning model after removing problematic training samples in cleaning or selecting different subsets of training data for interpretability. This paper presents an efficient provenance-based approach, PrIU, and its optimized version, PrIU-opt, for incrementally updating model parameters without sacrificing prediction accuracy. We prove the correctness and convergence of the incrementally updated model parameters, and validate it experimentally. Experimental results show that up to two orders of magnitude speed-ups can be achieved by PrIU-opt compared to simply retraining the model from scratch, yet obtaining highly similar models.
DOI: 10.1007/s00109-021-02166-z
发表时间: 2022-03
期刊: Journal of molecular medicine (Berlin, Germany)
影响因子: --
作者:
Im NR;Kim B;Jung KY;Baek SK
通讯作者: Baek SK
ORCHESTRA 协作数据共享系统
DOI: --
发表时间: 2008
期刊: SIGMOD record
影响因子: 1.1
作者:
Z. Ives;Todd J. Green;Gregory Karvounarakis;Nicholas E. Taylor;V. Tannen;P. Talukdar;Marie Jacob;Fernando C Pereira
通讯作者: Fernando C Pereira
DOI: 10.1007/s001900000125
发表时间: 2001
期刊: Journal of Geodesy
影响因子: 4.4
作者:
Wenke Sun;L. Sjöberg
通讯作者: L. Sjöberg
DOI: 10.1016/s0898-1221(03)90170-2
发表时间: 2003-05
期刊: --
影响因子: --
作者:
T. Dasu;T. Johnson
通讯作者: T. Dasu;T. Johnson
PALM:迭代调试的机器学习解释
DOI: --
发表时间: 2017
期刊: HILDA@SIGMOD
影响因子: --
作者:
S. Krishnan;Eugene Wu
通讯作者: Eugene Wu