Towards Better Performance in Incomplete LDL: Addressing Data Imbalance

TL;DR

This work proposes a framework that simultaneously handles incomplete labels and imbalanced label distributions, and decomposes the label distribution matrix into a low-rank component for frequent labels and a sparse component for rare labels, effectively capturing the structure of both head and tail labels.

Abstract

Label Distribution Learning (LDL) is a novel machine learning paradigm that addresses the problem of label ambiguity and has found widespread applications. Obtaining complete label distributions in real-world scenarios is challenging, which has led to the emergence of Incomplete Label Distribution Learning (InLDL). However, the existing InLDL methods overlook a crucial aspect of LDL data: the inherent imbalance in label distributions. To address this limitation, we propose \textbf{Incomplete and Imbalance Label Distribution Learning (I\(^2\)LDL)}, a framework that simultaneously handles incomplete labels and imbalanced label distributions. Our method decomposes the label distribution matrix into a low-rank component for frequent labels and a sparse component for rare labels, effectively capturing the structure of both head and tail labels. We optimize the model using the Alternating Direction Method of Multipliers (ADMM) and derive generalization error bounds via Rademacher complexity, providing strong theoretical guarantees. Extensive experiments on 15 real-world datasets demonstrate the effectiveness and robustness of our proposed framework compared to existing InLDL methods.

Figures (4)

• Figure 1: CD diagrams of the comparing algorithms in terms of each evaluation criterion. For the tests, CD equals 2.25 at 0.05 significance level.
• Figure 2: Comparison of different metrics across datasets for $I^2LDL$, $I^2LDL-a$, and $I^2LDL-b$ models.
• Figure 3: Convergence Curves for Different $\lambda$ Values
• Figure 4: Convergence Curves for Different Datasets