Sign-Guided Bipartite Graph Hashing for Hamming Space Search
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909189441323008 |
|---|---|
| author | Wu, Xueyi |
| author_facet | Wu, Xueyi |
| contents | Bipartite graph hashing (BGH) is extensively used for Top-K search in Hamming space at low storage and inference costs. Recent research adopts graph convolutional hashing for BGH and has achieved the state-of-the-art performance. However, the contributions of its various influencing factors to hashing performance have not been explored in-depth, including the same/different sign count between two binary embeddings during Hamming space search (sign property), the contribution of sub-embeddings at each layer (model property), the contribution of different node types in the bipartite graph (node property), and the combination of augmentation methods. In this work, we build a lightweight graph convolutional hashing model named LightGCH by mainly removing the augmentation methods of the state-of-the-art model BGCH. By analyzing the contributions of each layer and node type to performance, as well as analyzing the Hamming similarity statistics at each layer, we find that the actual neighbors in the bipartite graph tend to have low Hamming similarity at the shallow layer, and all nodes tend to have high Hamming similarity at the deep layers in LightGCH. To tackle these problems, we propose a novel sign-guided framework SGBGH to make improvement, which uses sign-guided negative sampling to improve the Hamming similarity of neighbors, and uses sign-aware contrastive learning to help nodes learn more uniform representations. Experimental results show that SGBGH outperforms BGCH and LightGCH significantly in embedding quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_02716 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Sign-Guided Bipartite Graph Hashing for Hamming Space Search Wu, Xueyi Information Retrieval Bipartite graph hashing (BGH) is extensively used for Top-K search in Hamming space at low storage and inference costs. Recent research adopts graph convolutional hashing for BGH and has achieved the state-of-the-art performance. However, the contributions of its various influencing factors to hashing performance have not been explored in-depth, including the same/different sign count between two binary embeddings during Hamming space search (sign property), the contribution of sub-embeddings at each layer (model property), the contribution of different node types in the bipartite graph (node property), and the combination of augmentation methods. In this work, we build a lightweight graph convolutional hashing model named LightGCH by mainly removing the augmentation methods of the state-of-the-art model BGCH. By analyzing the contributions of each layer and node type to performance, as well as analyzing the Hamming similarity statistics at each layer, we find that the actual neighbors in the bipartite graph tend to have low Hamming similarity at the shallow layer, and all nodes tend to have high Hamming similarity at the deep layers in LightGCH. To tackle these problems, we propose a novel sign-guided framework SGBGH to make improvement, which uses sign-guided negative sampling to improve the Hamming similarity of neighbors, and uses sign-aware contrastive learning to help nodes learn more uniform representations. Experimental results show that SGBGH outperforms BGCH and LightGCH significantly in embedding quality. |
| title | Sign-Guided Bipartite Graph Hashing for Hamming Space Search |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2405.02716 |