.png){kind=icon}
返回
Interpretable domain adaptation using unsupervised feature selection on pre-trained source models
.png){kind=icon}
OA
AI
DOI：10.1016/j.neucom.2022.09.031.png)
.png){kind=icon}
.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
We study a realistic domain adaptation setting where one has access to an already existing black-box machine learning model. Indeed, in real-life scenarios, an efficient pre-trained source domain predictive model is often available and required to be preserved. The solution we propose to this problem has the asset of providing an interpretable target to source transformation by seeking a sparse and ordered coordinate-wise adaptation of the feature space in addition to elementary mapping functions. To automat-ically select the subset of features to be adapted, we first introduce a weakly-supervised process relying on scarce labeled target data. Then, we address a more challenging unsupervised version of this domain adaptation scenario. To this end, we propose a new pseudo-label estimator over unlabeled target exam-ples, which is based on rank-stability in regards to the source model prediction. Such estimated labels are further used in a feature selection process to assess whether each feature needs to be transformed to achieve adaptation. We provide theoretical foundations of our method as well as an efficient implemen-tation. Numerical experiments on real datasets show particularly encouraging results since approaching the supervised case, where one has access to labeled target samples. (c) 2022 Elsevier B.V. All rights reserved.
Keyword:
Domain adaptation
Feature selection
Fraud detection
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
6.5
论文数:
2.5W
被引数:
6.5W
机构
引用论文
.png){kind=icon}
Enhanced Thermoelectric Performance of Bi2Te2.7Se0.3/Bi2S3 Synthesized by Anion Exchange Method
PREAI
分享收藏
查看更多内容