Meta Objective Guided Disambiguation for Partial Label Learning

TL;DR

This paper introduces MoGD, a novel partial label learning framework that uses a meta objective on a small validation set to effectively disambiguate candidate labels, improving performance over existing methods.

Contribution

The paper proposes a meta objective guided disambiguation framework for PLL that re-weights candidate labels based on validation loss, with proven convergence and error bounds.

Findings

Achieves superior performance on benchmark datasets.

Effectively disambiguates candidate labels using meta-guided re-weighting.

Proves convergence and derives error bounds for the method.

Abstract

Partial label learning (PLL) is a typical weakly supervised learning framework, where each training instance is associated with a candidate label set, among which only one label is valid. To solve PLL problems, typically methods try to perform disambiguation for candidate sets by either using prior knowledge, such as structure information of training data, or refining model outputs in a self-training manner. Unfortunately, these methods often fail to obtain a favorable performance due to the lack of prior information or unreliable predictions in the early stage of model training. In this paper, we propose a novel framework for partial label learning with meta objective guided disambiguation (MoGD), which aims to recover the ground-truth label from candidate labels set by solving a meta objective on a small validation set. Specifically, to alleviate the negative impact of false positive labels, we re-weight each candidate label based on the meta loss on the validation set. Then, the classifier is trained by minimizing the weighted cross entropy loss. The proposed method can be easily implemented by using various deep networks with the ordinary SGD optimizer. Theoretically, we prove the convergence property of meta objective and derive the estimation error bounds of the proposed method. Extensive experiments on various benchmark datasets and real-world PLL datasets demonstrate that the proposed method can achieve competent performance when compared with the state-of-the-art methods.