arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2106.12887cs.LGcs.AIstat.ML
收藏

一种消除训练好的机器学习模型偏差的近优算法

A Near-Optimal Algorithm for Debiasing Trained Machine Learning Models

Google Research(谷歌研究院)

Ibrahim Alabdulmohsin, Mario Lucic

摘要:

我们提出了一种可扩展的后处理算法,用于消除训练模型的偏差,包括深度神经网络(DNN),我们通过限制其超额贝叶斯风险来证明其接近最优。我们在经典算法和现代DNN架构的标准基准数据集上实证验证了它的优势,并证明它优于以前的后处理方法,同时与处理中的性能相当。此外,我们还证明了该算法对于大规模训练的模型特别有效,其中后处理是一种自然和实用的选择。

英文摘要:

We present a scalable post-processing algorithm for debiasing trained models, including deep neural networks (DNNs), which we prove to be near-optimal by bounding its excess Bayes risk. We empirically validate its advantages on standard benchmark datasets across both classical algorithms as well as modern DNN architectures and demonstrate that it outperforms previous post-processing methods while performing on par with in-processing. In addition, we show that the proposed algorithm is particularly effective for models trained at scale where post-processing is a natural and practical choice.

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。

↑