Back to Search
Start Over
Sign Rank Limitations for Inner Product Graph Decoders
- Publication Year :
- 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.<br />Comment: Accepted by ICML 2024
- Subjects :
- Computer Science - Machine Learning
Subjects
Details
- Database :
- arXiv
- Publication Type :
- Report
- Accession number :
- edsarx.2402.06662
- Document Type :
- Working Paper