Spectral Geometric Matrix Completion

Amit Boyarski, Sanketh Vedula, Alex Bronstein
Proceedings of the 2nd Mathematical and Scientific Machine Learning Conference, PMLR 145:172-196, 2022.

Abstract

Deep Matrix Factorization (DMF) is an emerging approach to the problem of matrix completion. Recent works have established that gradient descent applied to a DMF model induces an implicit regularization on the rank of the recovered matrix. In this work we interpret the DMF model through the lens of spectral geometry. This allows us to incorporate explicit regularization without breaking the DMF structure, thus enjoying the best of both worlds. In particular, we focus on matrix completion problems with underlying geometric or topological relations between the rows and/or columns. Such relations are prevalent in matrix completion problems that arise in many applications, such as recommender systems and drug-target interaction. Our contributions enable DMF models to exploit these relations, and make them competitive on real benchmarks, while exhibiting one of the first successful applications of deep linear networks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v145-boyarski22a, title = {Spectral Geometric Matrix Completion}, author = {Boyarski, Amit and Vedula, Sanketh and Bronstein, Alex}, booktitle = {Proceedings of the 2nd Mathematical and Scientific Machine Learning Conference}, pages = {172--196}, year = {2022}, editor = {Bruna, Joan and Hesthaven, Jan and Zdeborova, Lenka}, volume = {145}, series = {Proceedings of Machine Learning Research}, month = {16--19 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v145/boyarski22a/boyarski22a.pdf}, url = {https://proceedings.mlr.press/v145/boyarski22a.html}, abstract = { Deep Matrix Factorization (DMF) is an emerging approach to the problem of matrix completion. Recent works have established that gradient descent applied to a DMF model induces an implicit regularization on the rank of the recovered matrix. In this work we interpret the DMF model through the lens of spectral geometry. This allows us to incorporate explicit regularization without breaking the DMF structure, thus enjoying the best of both worlds. In particular, we focus on matrix completion problems with underlying geometric or topological relations between the rows and/or columns. Such relations are prevalent in matrix completion problems that arise in many applications, such as recommender systems and drug-target interaction. Our contributions enable DMF models to exploit these relations, and make them competitive on real benchmarks, while exhibiting one of the first successful applications of deep linear networks.} }
Endnote
%0 Conference Paper %T Spectral Geometric Matrix Completion %A Amit Boyarski %A Sanketh Vedula %A Alex Bronstein %B Proceedings of the 2nd Mathematical and Scientific Machine Learning Conference %C Proceedings of Machine Learning Research %D 2022 %E Joan Bruna %E Jan Hesthaven %E Lenka Zdeborova %F pmlr-v145-boyarski22a %I PMLR %P 172--196 %U https://proceedings.mlr.press/v145/boyarski22a.html %V 145 %X Deep Matrix Factorization (DMF) is an emerging approach to the problem of matrix completion. Recent works have established that gradient descent applied to a DMF model induces an implicit regularization on the rank of the recovered matrix. In this work we interpret the DMF model through the lens of spectral geometry. This allows us to incorporate explicit regularization without breaking the DMF structure, thus enjoying the best of both worlds. In particular, we focus on matrix completion problems with underlying geometric or topological relations between the rows and/or columns. Such relations are prevalent in matrix completion problems that arise in many applications, such as recommender systems and drug-target interaction. Our contributions enable DMF models to exploit these relations, and make them competitive on real benchmarks, while exhibiting one of the first successful applications of deep linear networks.
APA
Boyarski, A., Vedula, S. & Bronstein, A.. (2022). Spectral Geometric Matrix Completion. Proceedings of the 2nd Mathematical and Scientific Machine Learning Conference, in Proceedings of Machine Learning Research 145:172-196 Available from https://proceedings.mlr.press/v145/boyarski22a.html.

Related Material