Disentangling Degree-related Biases and Interest for Out-of-Distribution Generalized Directed Network Embedding
- Authors
- Yoo, Hyunsik; Lee, Yeon-Chang; Shin, Kijung; Kim, Sang-Wook
- Issue Date
- Apr-2023
- Publisher
- Association for Computing Machinery, Inc
- Keywords
- degree-related distributional shifts; network embedding
- Citation
- ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023, pp.231 - 239
- Indexed
- SCOPUS
- Journal Title
- ACM Web Conference 2023 - Proceedings of the World Wide Web Conference, WWW 2023
- Start Page
- 231
- End Page
- 239
- URI
- https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/185834
- DOI
- 10.1145/3543507.3583271
- Abstract
- The goal of directed network embedding is to represent the nodes in a given directed network as embeddings that preserve the asymmetric relationships between nodes. While a number of directed network embedding methods have been proposed, we empirically show that the existing methods lack out-of-distribution generalization abilities against degree-related distributional shifts. To mitigate this problem, we propose ODIN (Out-of-Distribution Generalized Directed Network Embedding), a new directed NE method where we model multiple factors in the formation of directed edges. Then, for each node, ODIN learns multiple embeddings, each of which preserves its corresponding factor, by disentangling interest factors and biases related to in- and out-degrees of nodes. Our experiments on four real-world directed networks demonstrate that disentangling multiple factors enables ODIN to yield out-of-distribution generalized embeddings that are consistently effective under various degrees of shifts in degree distributions. Specifically, ODIN universally outperforms 9 state-of-the-art competitors in 2 LP tasks on 4 real-world datasets under both identical distribution (ID) and non-ID settings. The code is available at https://github.com/hsyoo32/odin.
- Files in This Item
-
Go to Link
- Appears in
Collections - 서울 공과대학 > 서울 컴퓨터소프트웨어학부 > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/185834)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.