Title:
SYSTEM, METHOD, AND COMPUTER-READABLE MEDIA FOR LEAKAGE CORRECTION IN GRAPH NEURAL NETWORK BASED RECOMMENDER SYSTEMS
Document Type and Number:
WIPO Patent Application WO/2022/247878
Kind Code:
A1
Abstract:
Systems, methods, and computer-readable media provide a graph processing system that incorporates a graph neural network (GNN) based recommender system (RS), as well as a method for training a GNN based RS to address feature leakage that leads to overfitting of the trained GNN based RS. A message correction algorithm is used to modify a user node embedding and a positive item node embedding generated by the graph neural network when generating mini batches of training triples used to train the GNN based RS. The GNN message passing operations are performed on one graph only, in contrast to existing approaches which typically run GNN message passing operations on multiple adjusted input graphs constructed for multiple training triples.
Inventors:
KUMAR ISHAAN (CA)
HU YAOCHEN (CA)
ZHANG YINGXUE (CA)
HU YAOCHEN (CA)
ZHANG YINGXUE (CA)
Application Number:
PCT/CN2022/095106
Publication Date:
December 01, 2022
Filing Date:
May 26, 2022
Export Citation:
Assignee:
HUAWEI TECH CO LTD (CN)
International Classes:
G06N3/08
Foreign References:
| US20190080383A1 | 2019-03-14 | |||
| US20210051121A1 | 2021-02-18 | |||
| US20200250734A1 | 2020-08-06 | |||
| CN111143705A | 2020-05-12 |
Download PDF:
Previous Patent: AEROSOL GENERATING DEVICE
Next Patent: COMMUNICATION METHOD AND RELATED COMMUNICATION APPARATUS
Next Patent: COMMUNICATION METHOD AND RELATED COMMUNICATION APPARATUS