Advances in Weakly Supervised Learning: Theories, Algorithms, and Applications

Dear Colleagues,

Weakly supervised learning has emerged as a key area in machine learning that addresses the challenge of learning from imperfect, incomplete, or ambiguously labeled data. It encompasses a range of important paradigms, such as partial label learning, noisy label learning, incomplete multi-label learning, as well as semi-supervised and multi-instance learning. As real-world applications increasingly rely on large-scale and complex data, the need for learning methods that reduce annotation costs while maintaining model robustness has become more critical than ever. This Special Issue aims to gather cutting-edge research and new insights into the theories, methodologies, and applications of weakly supervised learning.

We welcome contributions that advance the foundations of weak supervision—such as risk-consistent estimators, disambiguation strategies, generalization analysis, and scalable algorithms—as well as those introducing novel practical frameworks, including deep weakly supervised models, integration with large language models (LLMs), multimodal and cross-modal weak supervision, and continual or federated learning under weak supervision. Studies that explore the connections between different weakly supervised paradigms (e.g., between partial label and noisy label learning) or enhance explainability, fairness, and adaptability under distribution shifts are also highly encouraged.

By bringing together diverse perspectives from both academia and industry, this Special Issue seeks to promote a holistic understanding of weakly supervised learning, inspire novel methodology development, and foster real-world deployments in domains such as computer vision, natural language processing, healthcare, recommendation systems, and scientific data analysis. We invite submissions related to, but not limited to, the following topics:

• Theoretical analyses and frameworks for weakly supervised learning.
• Deep architectures and optimization methods for weak supervision.
• The integration of weak supervision with foundation models and LLMs.
• Multi-modal and cross-modal weakly supervised learning.
• Continual, lifelong, and federated learning under weak supervision.
• Weakly supervised learning under distribution shifts and open-world settings.
• Connections and integrations between partial label, noisy label, multi-label, and other weakly supervised learning paradigms.
• Explainability, robustness, and uncertainty quantification in weak supervision.
• Real-world applications and benchmarking in the era of foundation models.

Dr. Beibei Li
Dr. Shuo He
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 250 words) can be sent to the Editorial Office for assessment.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Information is an international peer-reviewed open access monthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 1800 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.