Efficient Graph based Recommender System with Weighted Averaging of Messages
This work addresses a domain-specific challenge for Amazon sellers by providing an incremental improvement in efficiency and performance for graph-based recommender systems under cold start conditions.
The paper tackles the perpetual soft item cold start problem in Amazon's recommendation system for sellers by proposing a method that reduces graph size and uses a Weighted Averaging of Messages over Layers (WAML) algorithm, achieving a compute time reduction to 1/7 of LightGCN and 1/26 of GAT while increasing recall@100 by 66% over LightGCN and 2.3x over GAT.
We showcase a novel solution to a recommendation system problem where we face a perpetual soft item cold start issue. Our system aims to recommend demanded products to prospective sellers for listing in Amazon stores. These products always have only few interactions thereby giving rise to a perpetual soft item cold start situation. Modern collaborative filtering methods solve cold start using content attributes and exploit the existing implicit signals from warm start items. This approach fails in our use-case since our entire item set faces cold start issue always. Our Product Graph has over 500 Million nodes and over 5 Billion edges which makes training and inference using modern graph algorithms very compute intensive. To overcome these challenges we propose a system which reduces the dataset size and employs an improved modelling technique to reduce storage and compute without loss in performance. Particularly, we reduce our graph size using a filtering technique and then exploit this reduced product graph using Weighted Averaging of Messages over Layers (WAML) algorithm. WAML simplifies training on large graphs and improves over previous methods by reducing compute time to 1/7 of LightGCN and 1/26 of Graph Attention Network (GAT) and increasing recall$@100$ by 66% over LightGCN and 2.3x over GAT.