FastBUS: A Fast Bayesian Framework for Unified Weakly-Supervised Learning

2026-02-28

AI Summary

Existing general-purpose weakly supervised learning methods often suffer from complex preprocessing, neglect of label dependencies, difficulty in batch processing, and low computational efficiency. This work proposes a unified Bayesian framework that, for the first time, compresses diverse weak supervision signals into a shared Bayesian network. By modeling label search as a probabilistic transition process and leveraging generalized belief propagation to infer the underlying true label distribution, the approach achieves both expressiveness and scalability. The method innovatively incorporates low-rank approximation and an end-to-end state evolution module to enable efficient batch inference, and it establishes a theoretical equivalence to the Expectation-Maximization (EM) algorithm. Extensive experiments demonstrate state-of-the-art performance across multiple weak supervision settings, with runtime improvements of up to several hundred times over current general-purpose baselines.

Technology Category

• Machine Learning: Bayesian Learning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Search and Optimization: Learning to Search

Application Category

• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Machine Learning often involves various imprecise labels, leading to diverse weakly supervised settings. While recent methods aim for universal handling, they usually suffer from complex manual pre-work, ignore the relationships between associated labels, or are unable to batch process due to computational design flaws, resulting in long running times. To address these limitations, we propose a novel general framework that efficiently infers latent true label distributions across various weak supervisions. Our key idea is to express the label brute-force search process as a probabilistic transition of label variables, compressing diverse weakly supervised DFS tree structures into a shared Bayesian network. From this, we derived a latent probability calculation algorithm based on generalized belief propagation and proposed two joint acceleration strategies: 1) introducing a low-rank assumption to approximate the transition matrix, reducing time complexity; 2) designing an end-to-end state evolution module to learn batch-scale transition matrices, facilitating multi-category batch processing. In addition, the equivalence of our method with the EM algorithm in most scenarios is further demonstrated. Extensive experiments show that our method achieves SOTA results under most weakly supervised settings, and achieves up to hundreds of times faster acceleration in running time compared to other general methods.

Problem

• weakly-supervised learning
• label imprecision
• computational efficiency
• batch processing
• latent label inference

Innovation

• Bayesian network
• weakly-supervised learning
• belief propagation
• low-rank approximation
• batch processing