Network Embedding as Matrix Factorization: Unifying DeepWalk, LINE, PTE, and node2vec · HackerPapers discussions

Network Embedding as Matrix Factorization: Unifying DeepWalk, LINE, PTE, and node2vec — mentioned by khelavastr

Tue, 18 Feb 2025 00:17:42 GMT

Have you seen, "Network Embedding as Matrix Factorization: Unifying DeepWalk, LINE, PTE, and node2vec"..? You can node2vec any symmetric graph to approximate transforms of its eigendecomposition. https://arxiv.org/abs/1710.02971

Paper and discussion history. A mention is not an endorsement.

Network Embedding as Matrix Factorization: Unifying DeepWalk, LINE, PTE, and node2vec — mentioned by Khelavaster

Wed, 22 May 2024 00:15:05 GMT

More accurately, vector databases are Softmax'd eigendecompositions of the matrix representation of the graph of their connected members.. "Network Embedding as Matrix Factorization: Unifying DeepWalk, LINE, PTE, and node2vec" https://arxiv.org/abs/1710.02971

Paper and discussion history. A mention is not an endorsement.

Network Embedding as Matrix Factorization: Unifying DeepWalk, LINE, PTE, and node2vec — HN submission

Tue, 25 Aug 2020 22:25:01 GMT

Network Embedding as Matrix Factorization: DeepWalk, Line, PTE, Node2vec Unified

2 points; 0 discussion replies on HN. Paper and source history. A submission is not an endorsement.