We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.LG

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Machine Learning

Title: Efficient and effective training of language and graph neural network models

Abstract: Can we combine heterogenous graph structure with text to learn high-quality semantic and behavioural representations? Graph neural networks (GNN)s encode numerical node attributes and graph structure to achieve impressive performance in a variety of supervised learning tasks. Current GNN approaches are challenged by textual features, which typically need to be encoded to a numerical vector before provided to the GNN that may incur some information loss. In this paper, we put forth an efficient and effective framework termed language model GNN (LM-GNN) to jointly train large-scale language models and graph neural networks. The effectiveness in our framework is achieved by applying stage-wise fine-tuning of the BERT model first with heterogenous graph information and then with a GNN model. Several system and design optimizations are proposed to enable scalable and efficient training. LM-GNN accommodates node and edge classification as well as link prediction tasks. We evaluate the LM-GNN framework in different datasets performance and showcase the effectiveness of the proposed approach. LM-GNN provides competitive results in an Amazon query-purchase-product application.
Subjects: Machine Learning (cs.LG); Computation and Language (cs.CL)
Cite as: arXiv:2206.10781 [cs.LG]
  (or arXiv:2206.10781v1 [cs.LG] for this version)

Submission history

From: Vassilis N. Ioannidis [view email]
[v1] Wed, 22 Jun 2022 00:23:37 GMT (109kb)

Link back to: arXiv, form interface, contact.