Abstract
Present disclosure discloses a computer-implemented method of positional embedding for encoding positional information. The computer-implemented method includes generating a positional embedding for a node utilizing a random walk with restart (RWR) proximity distribution to generate a pairwise proximity distribution matrix comprising a plurality of rows. The computer-implemented method includes sorting the pairwise proximity distribution matrix for the node. The computer-implemented method includes generating a structural embedding based on the sorting of the pairwise proximity distribution matrix for the node.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Chen, Yuzhong; Pan, Menghai; Chen, Huiyuan; Das, Mahashweta; and Yang, Hao, "POSITIONAL EMBEDDING MEETS STRUCTURAL EMBEDDING OF GRAPH REPRESENTATION LEARNING", Technical Disclosure Commons, (April 12, 2024)
https://www.tdcommons.org/dpubs_series/6871