Hierarchical Graph Information Bottleneck for Multi-Behavior Recommendation

2025-07-21

AI Summary

In multi-behavior recommendation, auxiliary behaviors (e.g., clicks, views) exhibit significant distributional divergence from the target behavior (e.g., purchases) and introduce noise, leading to negative transfer. To address this, we propose a model-agnostic hierarchical graph information bottleneck framework. Our method employs a dynamic edge-pruning graph refinement encoder to construct compact and sufficient user representations, integrating multi-behavior relational modeling with representation compression within a hierarchical graph neural network. By explicitly incorporating the information bottleneck principle, the framework suppresses noise from auxiliary behaviors, mitigates negative transfer, and enhances cross-behavior knowledge transfer efficiency. Extensive evaluations on three public benchmarks and multiple industrial scenarios demonstrate substantial improvements in offline metrics; online A/B tests show an average 2.3% lift in conversion rate.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Graph-based Machine Learning
• Cognitive Modeling & Cognitive Systems: Adaptive Behavior

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

In real-world recommendation scenarios, users typically engage with platforms through multiple types of behavioral interactions. Multi-behavior recommendation algorithms aim to leverage various auxiliary user behaviors to enhance prediction for target behaviors of primary interest (e.g., buy), thereby overcoming performance limitations caused by data sparsity in target behavior records. Current state-of-the-art approaches typically employ hierarchical design following either cascading (e.g., view$ ightarrow$cart$ ightarrow$buy) or parallel (unified$ ightarrow$behavior$ ightarrow$specific components) paradigms, to capture behavioral relationships. However, these methods still face two critical challenges: (1) severe distribution disparities across behaviors, and (2) negative transfer effects caused by noise in auxiliary behaviors. In this paper, we propose a novel model-agnostic Hierarchical Graph Information Bottleneck (HGIB) framework for multi-behavior recommendation to effectively address these challenges. Following information bottleneck principles, our framework optimizes the learning of compact yet sufficient representations that preserve essential information for target behavior prediction while eliminating task-irrelevant redundancies. To further mitigate interaction noise, we introduce a Graph Refinement Encoder (GRE) that dynamically prunes redundant edges through learnable edge dropout mechanisms. We conduct comprehensive experiments on three real-world public datasets, which demonstrate the superior effectiveness of our framework. Beyond these widely used datasets in the academic community, we further expand our evaluation on several real industrial scenarios and conduct an online A/B testing, showing again a significant improvement in multi-behavior recommendations. The source code of our proposed HGIB is available at https://github.com/zhy99426/HGIB.

Problem

• Addresses data sparsity in target behavior records
• Mitigates distribution disparities across user behaviors
• Reduces noise-induced negative transfer effects

Innovation

• Hierarchical Graph Information Bottleneck framework
• Graph Refinement Encoder prunes edges
• Model-agnostic approach for multi-behavior recommendation