Inaccurate Label Distribution Learning with Dependency Noise

TL;DR

This paper presents DN-ILDL, a novel framework that models and mitigates dependency noise in label distribution learning, improving accuracy through noise decomposition, graph regularization, and efficient optimization.

Contribution

The paper introduces a dependency noise-aware framework for label distribution learning, incorporating noise modeling, matrix decomposition, graph regularization, and ADMM optimization.

Findings

DN-ILDL outperforms existing LDL methods in experiments.

The framework effectively recovers true label distributions.

The method provides a generalization error bound.

Abstract

In this paper, we introduce the Dependent Noise-based Inaccurate Label Distribution Learning (DN-ILDL) framework to tackle the challenges posed by noise in label distribution learning, which arise from dependencies on instances and labels. We start by modeling the inaccurate label distribution matrix as a combination of the true label distribution and a noise matrix influenced by specific instances and labels. To address this, we develop a linear mapping from instances to their true label distributions, incorporating label correlations, and decompose the noise matrix using feature and label representations, applying group sparsity constraints to accurately capture the noise. Furthermore, we employ graph regularization to align the topological structures of the input and output spaces, ensuring accurate reconstruction of the true label distribution matrix. Utilizing the Alternating Direction Method of Multipliers (ADMM) for efficient optimization, we validate our method's capability to recover true labels accurately and establish a generalization error bound. Extensive experiments demonstrate that DN-ILDL effectively addresses the ILDL problem and outperforms existing LDL methods.