Amortized Variational Inference for Partial-Label Learning: A Probabilistic Approach to Label Disambiguation

TL;DR

This paper introduces a probabilistic framework using amortized variational inference for partial-label learning, effectively disambiguating labels in noisy, ambiguous data with improved accuracy and computational efficiency.

Contribution

It presents a novel neural network-based variational inference method for partial-label learning, combining deep learning expressiveness with probabilistic rigor, and demonstrating state-of-the-art results.

Findings

Achieves superior accuracy on synthetic and real datasets.

Offers significant computational efficiency improvements.

Maintains flexibility across different neural network architectures.

Abstract

Real-world data is frequently noisy and ambiguous. In crowdsourcing, for example, human annotators may assign conflicting class labels to the same instances. Partial-label learning (PLL) addresses this challenge by training classifiers when each instance is associated with a set of candidate labels, only one of which is correct. While early PLL methods approximate the true label posterior, they are often computationally intensive. Recent deep learning approaches improve scalability but rely on surrogate losses and heuristic label refinement. We introduce a novel probabilistic framework that directly approximates the posterior distribution over true labels using amortized variational inference. Our method employs neural networks to predict variational parameters from input data, enabling efficient inference. This approach combines the expressiveness of deep learning with the rigor of probabilistic modeling, while remaining architecture-agnostic. Theoretical analysis and extensive experiments on synthetic and real-world datasets demonstrate that our method achieves state-of-the-art performance in both accuracy and efficiency.