References & Citations
Computer Science > Machine Learning
Title: Partial Label Learning with a Partner
(Submitted on 18 Dec 2023 (this version), latest version 28 Mar 2024 (v3))
Abstract: In partial label learning (PLL), each instance is associated with a set of candidate labels among which only one is ground-truth. The majority of the existing works focuses on constructing robust classifiers to estimate the labeling confidence of candidate labels in order to identify the correct one. However, these methods usually struggle to rectify mislabeled samples. To help existing PLL methods identify and rectify mislabeled samples, in this paper, we introduce a novel partner classifier and propose a novel ``mutual supervision'' paradigm. Specifically, we instantiate the partner classifier predicated on the implicit fact that non-candidate labels of a sample should not be assigned to it, which is inherently accurate and has not been fully investigated in PLL. Furthermore, a novel collaborative term is formulated to link the base classifier and the partner one. During each stage of mutual supervision, both classifiers will blur each other's predictions through a blurring mechanism to prevent overconfidence in a specific label. Extensive experiments demonstrate that the performance and disambiguation ability of several well-established stand-alone and deep-learning based PLL approaches can be significantly improved by coupling with this learning paradigm.
Submission history
From: Chongjie Si [view email][v1] Mon, 18 Dec 2023 09:09:52 GMT (2071kb,D)
[v2] Thu, 18 Jan 2024 05:20:27 GMT (2065kb,D)
[v3] Thu, 28 Mar 2024 04:46:19 GMT (5212kb,D)
Link back to: arXiv, form interface, contact.