Multi-Tower Multi-Interest Recommendation with User Representation Repel
This work addresses key bottlenecks in multi-interest learning for recommender systems, offering a solution that improves performance and adoptability in industrial settings, though it appears incremental as it builds on existing multi-interest methods.
The paper tackled the challenges of multi-interest sequential recommendation, including training-deployment objective mismatch, lack of item information access, and industrial adoption difficulties, by proposing a multi-tower multi-interest framework with user representation repel, achieving proven effectiveness and generalizability across large-scale industrial datasets.
In the era of information overload, the value of recommender systems has been profoundly recognized in academia and industry alike. Multi-interest sequential recommendation, in particular, is a subfield that has been receiving increasing attention in recent years. By generating multiple-user representations, multi-interest learning models demonstrate superior expressiveness than single-user representation models, both theoretically and empirically. Despite major advancements in the field, three major issues continue to plague the performance and adoptability of multi-interest learning methods, the difference between training and deployment objectives, the inability to access item information, and the difficulty of industrial adoption due to its single-tower architecture. We address these challenges by proposing a novel multi-tower multi-interest framework with user representation repel. Experimental results across multiple large-scale industrial datasets proved the effectiveness and generalizability of our proposed framework.