Deep Causal Metric Learning

Xiang Deng, Zhongfei Zhang
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:4993-5006, 2022.

Abstract

Deep metric learning aims to learn distance metrics that measure similarities and dissimilarities between samples. The existing approaches typically focus on designing different hard sample mining or distance margin strategies and then minimize a pair/triplet-based or proxy-based loss over the training data. However, this can lead the model to recklessly learn all the correlated distances found in training data including the spurious distance (e.g., background differences) that is not the distance of interest and can harm the generalization of the learned metric. To address this issue, we study metric learning from a causality perspective and accordingly propose deep causal metric learning (DCML) that pursues the true causality of the distance between samples. DCML is achieved through explicitly learning environment-invariant attention and task-invariant embedding based on causal inference. Extensive experiments on several benchmark datasets demonstrate the superiority of DCML over the existing methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v162-deng22c, title = {Deep Causal Metric Learning}, author = {Deng, Xiang and Zhang, Zhongfei}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {4993--5006}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/deng22c/deng22c.pdf}, url = {https://proceedings.mlr.press/v162/deng22c.html}, abstract = {Deep metric learning aims to learn distance metrics that measure similarities and dissimilarities between samples. The existing approaches typically focus on designing different hard sample mining or distance margin strategies and then minimize a pair/triplet-based or proxy-based loss over the training data. However, this can lead the model to recklessly learn all the correlated distances found in training data including the spurious distance (e.g., background differences) that is not the distance of interest and can harm the generalization of the learned metric. To address this issue, we study metric learning from a causality perspective and accordingly propose deep causal metric learning (DCML) that pursues the true causality of the distance between samples. DCML is achieved through explicitly learning environment-invariant attention and task-invariant embedding based on causal inference. Extensive experiments on several benchmark datasets demonstrate the superiority of DCML over the existing methods.} }
Endnote
%0 Conference Paper %T Deep Causal Metric Learning %A Xiang Deng %A Zhongfei Zhang %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-deng22c %I PMLR %P 4993--5006 %U https://proceedings.mlr.press/v162/deng22c.html %V 162 %X Deep metric learning aims to learn distance metrics that measure similarities and dissimilarities between samples. The existing approaches typically focus on designing different hard sample mining or distance margin strategies and then minimize a pair/triplet-based or proxy-based loss over the training data. However, this can lead the model to recklessly learn all the correlated distances found in training data including the spurious distance (e.g., background differences) that is not the distance of interest and can harm the generalization of the learned metric. To address this issue, we study metric learning from a causality perspective and accordingly propose deep causal metric learning (DCML) that pursues the true causality of the distance between samples. DCML is achieved through explicitly learning environment-invariant attention and task-invariant embedding based on causal inference. Extensive experiments on several benchmark datasets demonstrate the superiority of DCML over the existing methods.
APA
Deng, X. & Zhang, Z.. (2022). Deep Causal Metric Learning. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:4993-5006 Available from https://proceedings.mlr.press/v162/deng22c.html.

Related Material