Fast Graph Learning with Unique Optimal Solutions

Code

• github.com/samihaija/tf-fsvd
  OfficialIn papertf★ 13

Abstract

We consider two popular Graph Representation Learning (GRL) methods: message passing for node classification and network embedding for link prediction. For each, we pick a popular model that we: (i) linearize and (ii) and switch its training objective to Frobenius norm error minimization. These simplifications can cast the training into finding the optimal parameters in closed-form. We program in TensorFlow a functional form of Truncated Singular Value Decomposition (SVD), such that, we could decompose a dense matrix M, without explicitly computing M. We achieve competitive performance on popular GRL tasks while providing orders of magnitude speedup. We open-source our code at http://github.com/samihaija/tf-fsvd

Tasks

Reproductions