ZOOMER: Boosting Retrieval on Web-scale Graphs by Regions of Interest

TL;DR

ZOOMER is a system that improves GNN-based recommendations on large-scale graphs by focusing on Regions of Interest, reducing computational costs and enhancing recommendation quality at Taobao.

Contribution

Introduces the concept of Regions of Interest (ROI) in GNNs for recommendations, enabling efficient training and serving on web-scale graphs with improved accuracy.

Findings

Achieves up to 14x speedup in training and serving.

Maintains or improves AUC performance compared to baseline methods.

Effectively filters relevant user interests for better recommendations.

Abstract

We introduce ZOOMER, a system deployed at Taobao, the largest e-commerce platform in China, for training and serving GNN-based recommendations over web-scale graphs. ZOOMER is designed for tackling two challenges presented by the massive user data at Taobao: low training/serving efficiency due to the huge scale of the graphs, and low recommendation quality due to the information overload which distracts the recommendation model from specific user intentions. ZOOMER achieves this by introducing a key concept, Region of Interests (ROI) in GNNs for recommendations, i.e., a neighborhood region in the graph with significant relevance to a strong user intention. ZOOMER narrows the focus from the whole graph and "zooms in" on the more relevant ROIs, thereby reducing the training/serving cost and mitigating the information overload at the same time. With carefully designed mechanisms, ZOOMER identifies the interest expressed by each recommendation request, constructs an ROI subgraph by sampling with respect to the interest, and guides the GNN to reweigh different parts of the ROI towards the interest by a multi-level attention module. Deployed as a large-scale distributed system, ZOOMER supports graphs with billions of nodes for training and thousands of requests per second for serving. ZOOMER achieves up to 14x speedup when downsizing sampling scales with comparable (even better) AUC performance than baseline methods. Besides, both the offline evaluation and online A/B test demonstrate the effectiveness of ZOOMER.