29
0
0
2025-06-10
Special Issue on Beneficial Noise Learning Submission Date: 2025-06-15 Noise is an emerging and popular keyword in recent years. The noise-based models have attracted more and more attention in the pattern recognition community, including but not limited to random forest, dropout in neural networks (a kind of structural beneficial noise), generative adversarial networks, adversarial training, noisy augmentation, noisy label, positive-incentive noise, diffusion models, and flow matching models. Although most of these models don’t explicitly claim that they aim to learn noise, they actually utilize the beneficial noise implicitly. In many current studies, it is pointed out that noise can also be beneficial to large models. Therefore, noise should not be simply regarded as a harmful component any more. The positivity of noise deserves more systematic studies.
Although there are plenty of noise-related models, scientific studies of beneficial noise learning are still lacking to some extent. Most of these noise-based models just use beneficial noise in a heuristic way. This Special Issue calls for papers that study several attractive, natural, and urgent questions: (1) how a model learns the beneficial noise in a controllable manner; (2) what kind of noise will be beneficial to specific models/tasks; (3) the theoretical bound of beneficial noise.
This Special Issue seeks to cover a wide range of topics related to beneficial noise learning and analysis, including but not limited to:
Noise-based generative models such as GAN, diffusion models, and flow matching;
Beneficial noisy and uncertain structure in deep learning models;
Noisy model training such as beneficial noisy labels and adversarial training;
Noisy augmentations in diverse fields such as representation learning and signal detection;
Positive-incentive noise;
Beneficial noise in large models;
Beneficial noise in data acquisition;
Guest editors:
Xuelong Li, PhD
China Telecom, Beijing, China
Email: [email protected]
Hongyuan Zhang, PhD
The University of Hong Kong, HK, China
Email: [email protected]
Murat Sensoy, PhD
Amazon, Artificial General Intelligence (AGI) Department, London, UK
Email: [email protected]
Enze Xie, PhD
NVIDIA, Santa Clara, USA
Email: [email protected]
Manuscript submission information:
The journal submission system (Editorial Manager?) will be open for submissions to our Special Issue from November 15, 2024. When submitting your manuscript please select the article type VSI: Beneficial Noise Learning. Both the Guide for Authors and the submission portal could be found on the Journal Homepage: Guide for authors - Pattern Recognition - ISSN 0031-3203 | ScienceDirect.com by Elsevier.
Important dates
Submission Portal Open: November 15, 2024
Submission Deadline: June 15, 2025
Acceptance Deadline: August 15, 2025
Keywords:
Noise Learning, Beneficial Noise, Information Theory, Explainability, Generative Models, Uncertainty
