Label Distribution Learning Forests

TL;DR

This paper introduces Label Distribution Learning Forests (LDLFs), a novel approach using differentiable decision trees that model complex label distributions and integrate with representation learning, demonstrating superior performance on various tasks.

Contribution

The paper proposes LDLFs, a new LDL method based on differentiable decision trees that can model any label distribution form and enable joint learning with feature representations.

Findings

Significant improvements over state-of-the-art LDL methods.

Effective modeling of complex label distributions.

Successful application in computer vision tasks.

Abstract

Label distribution learning (LDL) is a general learning framework, which assigns to an instance a distribution over a set of labels rather than a single label or multiple labels. Current LDL methods have either restricted assumptions on the expression form of the label distribution or limitations in representation learning, e.g., to learn deep features in an end-to-end manner. This paper presents label distribution learning forests (LDLFs) - a novel label distribution learning algorithm based on differentiable decision trees, which have several advantages: 1) Decision trees have the potential to model any general form of label distributions by a mixture of leaf node predictions. 2) The learning of differentiable decision trees can be combined with representation learning. We define a distribution-based loss function for a forest, enabling all the trees to be learned jointly, and show that an update function for leaf node predictions, which guarantees a strict decrease of the loss function, can be derived by variational bounding. The effectiveness of the proposed LDLFs is verified on several LDL tasks and a computer vision application, showing significant improvements to the state-of-the-art LDL methods.