Addressing Missing Data Issue for Diffusion-based Recommendation

2025-05-18

AI Summary

To address the degradation of diffusion-based recommendation performance caused by missing data in user behavioral sequences, this paper proposes Thompson Sampling–driven Diffusion Models (TDM). Methodologically, TDM introduces a novel bilateral Thompson sampling mechanism that jointly models preference evolution along two dimensions: item continuity and sequence stability. It further incorporates a missingness-aware consistency regularization training paradigm, integrated with DDIM to enhance denoising robustness. Theoretically, TDM is proven to guarantee convergence under missing-data conditions. Empirically, it achieves significant improvements in recommendation accuracy and robustness across multiple public benchmarks, while accelerating inference by 3.2×. The core contributions are threefold: (i) the first integration of Thompson sampling into diffusion-based recommendation frameworks; (ii) synergistic modeling of active missing-data simulation and preference evolution; and (iii) a theoretically grounded, empirically validated approach for robust sequential recommendation under data sparsity.

Technology Category

• Data Mining & Knowledge Management: Recommender Systems
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Sequential Decision Making

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Web Mining and Content Analysis: Content-based information diffusion
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Diffusion models have shown significant potential in generating oracle items that best match user preference with guidance from user historical interaction sequences. However, the quality of guidance is often compromised by unpredictable missing data in observed sequence, leading to suboptimal item generation. Since missing data is uncertain in both occurrence and content, recovering it is impractical and may introduce additional errors. To tackle this challenge, we propose a novel dual-side Thompson sampling-based Diffusion Model (TDM), which simulates extra missing data in the guidance signals and allows diffusion models to handle existing missing data through extrapolation. To preserve user preference evolution in sequences despite extra missing data, we introduce Dual-side Thompson Sampling to implement simulation with two probability models, sampling by exploiting user preference from both item continuity and sequence stability. TDM strategically removes items from sequences based on dual-side Thompson sampling and treats these edited sequences as guidance for diffusion models, enhancing models' robustness to missing data through consistency regularization. Additionally, to enhance the generation efficiency, TDM is implemented under the denoising diffusion implicit models to accelerate the reverse process. Extensive experiments and theoretical analysis validate the effectiveness of TDM in addressing missing data in sequential recommendations.

Problem

• Handling unpredictable missing data in user interaction sequences
• Improving robustness of diffusion models for recommendation systems
• Enhancing generation efficiency with denoising diffusion implicit models

Innovation

• Dual-side Thompson sampling simulates missing data
• Extrapolation handles existing missing data issues
• Denoising diffusion implicit models accelerate generation