Partial-label Learning with Mixed Closed-set and Open-set Out-of-candidate Examples

TL;DR

This paper introduces a novel partial-label learning framework that effectively handles out-of-candidate examples, including both closed-set and open-set types, improving learning accuracy in complex real-world scenarios.

Contribution

The study pioneers a method to learn from out-of-candidate examples in partial-label learning by differentiating and leveraging closed-set and open-set OOC instances.

Findings

The proposed method outperforms existing PLL approaches in experiments.

Effective differentiation of closed-set and open-set OOC examples enhances learning.

Dynamic label assignment improves model robustness against OOC examples.

Abstract

Partial-label learning (PLL) relies on a key assumption that the true label of each training example must be in the candidate label set. This restrictive assumption may be violated in complex real-world scenarios, and thus the true label of some collected examples could be unexpectedly outside the assigned candidate label set. In this paper, we term the examples whose true label is outside the candidate label set OOC (out-of-candidate) examples, and pioneer a new PLL study to learn with OOC examples. We consider two types of OOC examples in reality, i.e., the closed-set/open-set OOC examples whose true label is inside/outside the known label space. To solve this new PLL problem, we first calculate the wooden cross-entropy loss from candidate and non-candidate labels respectively, and dynamically differentiate the two types of OOC examples based on specially designed criteria. Then, for closed-set OOC examples, we conduct reversed label disambiguation in the non-candidate label set; for open-set OOC examples, we leverage them for training by utilizing an effective regularization strategy that dynamically assigns random candidate labels from the candidate label set. In this way, the two types of OOC examples can be differentiated and further leveraged for model training. Extensive experiments demonstrate that our proposed method outperforms state-of-the-art PLL methods.