Sign Rank Limitations for Inner Product Graph Decoders

Su Hyeong Lee, Qingqi Zhang, Risi Kondor
Proceedings of the 41st International Conference on Machine Learning, PMLR 235:27118-27136, 2024.

Abstract

Inner product-based decoders are among the most influential frameworks used to extract meaningful data from latent embeddings. However, such decoders have shown limitations in representation capacity in numerous works within the literature, which have been particularly notable in graph reconstruction problems. In this paper, we provide the first theoretical elucidation of this pervasive phenomenon in graph data, and suggest straightforward modifications to circumvent this issue without deviating from the inner product framework.

Cite this Paper

BibTeX
@InProceedings{pmlr-v235-lee24ad, title = {Sign Rank Limitations for Inner Product Graph Decoders}, author = {Lee, Su Hyeong and Zhang, Qingqi and Kondor, Risi}, booktitle = {Proceedings of the 41st International Conference on Machine Learning}, pages = {27118--27136}, year = {2024}, editor = {Salakhutdinov, Ruslan and Kolter, Zico and Heller, Katherine and Weller, Adrian and Oliver, Nuria and Scarlett, Jonathan and Berkenkamp, Felix}, volume = {235}, series = {Proceedings of Machine Learning Research}, month = {21--27 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v235/main/assets/lee24ad/lee24ad.pdf}, url = {https://proceedings.mlr.press/v235/lee24ad.html}, abstract = {Inner product-based decoders are among the most influential frameworks used to extract meaningful data from latent embeddings. However, such decoders have shown limitations in representation capacity in numerous works within the literature, which have been particularly notable in graph reconstruction problems. In this paper, we provide the first theoretical elucidation of this pervasive phenomenon in graph data, and suggest straightforward modifications to circumvent this issue without deviating from the inner product framework.} }
Endnote
%0 Conference Paper %T Sign Rank Limitations for Inner Product Graph Decoders %A Su Hyeong Lee %A Qingqi Zhang %A Risi Kondor %B Proceedings of the 41st International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2024 %E Ruslan Salakhutdinov %E Zico Kolter %E Katherine Heller %E Adrian Weller %E Nuria Oliver %E Jonathan Scarlett %E Felix Berkenkamp %F pmlr-v235-lee24ad %I PMLR %P 27118--27136 %U https://proceedings.mlr.press/v235/lee24ad.html %V 235 %X Inner product-based decoders are among the most influential frameworks used to extract meaningful data from latent embeddings. However, such decoders have shown limitations in representation capacity in numerous works within the literature, which have been particularly notable in graph reconstruction problems. In this paper, we provide the first theoretical elucidation of this pervasive phenomenon in graph data, and suggest straightforward modifications to circumvent this issue without deviating from the inner product framework.
APA
Lee, S.H., Zhang, Q. & Kondor, R.. (2024). Sign Rank Limitations for Inner Product Graph Decoders. Proceedings of the 41st International Conference on Machine Learning, in Proceedings of Machine Learning Research 235:27118-27136 Available from https://proceedings.mlr.press/v235/lee24ad.html.

Related Material