Autoregressive Generation Strategies for Top-K Sequential Recommendations

Abstract

The goal of modern sequential recommender systems is often formulated in terms of next-item prediction. In this paper, we explore the applicability of generative transformer-based models for the Top-K sequential recommendation task, where the goal is to predict items a user is likely to interact with in the "near future". We explore commonly used autoregressive generation strategies, including greedy decoding, beam search, and temperature sampling, to evaluate their performance for the Top-K sequential recommendation task. In addition, we propose novel Reciprocal Rank Aggregation (RRA) and Relevance Aggregation (RA) generation strategies based on multi-sequence generation with temperature sampling and subsequent aggregation. Experiments on diverse datasets give valuable insights regarding commonly used strategies' applicability and show that suggested approaches improve performance on longer time horizons compared to widely-used Top-K prediction approach and single-sequence autoregressive generation strategies.

References in corpus (12)

• Contrastive Learning for Representation Degeneration Problem in Sequential Recommendation
• BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer
• Personalized Bundle List Recommendation
• MEANTIME: Mixture of Attention Mechanisms with Multi-temporal Embeddings for Sequential Recommendation
• Turning Dross Into Gold Loss: is BERT4Rec really better than SASRec?
• Widespread Flaws in Offline Evaluation of Recommender Systems
• Does It Look Sequential? An Analysis of Datasets for Evaluation of Sequential Recommendations
• Scaling Session-Based Transformer Recommendations using Optimized Negative Sampling and Loss Functions
• Scalable Cross-Entropy Loss for Sequential Recommendations with Large Item Catalogs
• Integrating Item Relevance in Training Loss for Sequential Recommender Systems
• RECE: Reduced Cross-Entropy Loss for Large-Catalogue Sequential Recommenders
• RePlay: a Recommendation Framework for Experimentation and Production Use