RADS: Reinforcement Learning-Based Sample Selection Improves Transfer Learning in Low-resource and Imbalanced Clinical Settings

Wei Han; David Martinez; Anna Khanina; Lawrence Cavedon; Karin Verspoor

cs.CLApr 22, 2026

RADS: Reinforcement Learning-Based Sample Selection Improves Transfer Learning in Low-resource and Imbalanced Clinical Settings

Authors: Wei Han, David Martinez, Anna Khanina, Lawrence Cavedon, Karin Verspoor

Organizations: School of Computing Technologies, RMIT University · Department of Infectious Disease, Peter MacCallum Cancer Centre · 3National Centre for Infections in Cancer, Melbourne · 5Sir Peter MacCallum Department of Oncology, The University of Melbourne · School of Computing and Information Systems, The University of Melbourne

Abstract

A common strategy in transfer learning is few shot fine-tuning, but its success is highly dependent on the quality of samples selected as training examples. Active learning methods such as uncertainty sampling and diversity sampling can select useful samples. However, under extremely low-resource and class-imbalanced conditions, they often favor outliers rather than truly informative samples, resulting in degraded performance. In this paper, we introduce RADS (Reinforcement Adaptive Domain Sampling), a robust sample selection strategy using reinforcement learning (RL) to identify the most informative samples. Experimental evaluations on several real world clinical datasets show our sample selection strategy enhances model transferability while maintaining robust performance under extreme class imbalance compared to traditional methods.

Explore similar work

Active few-shot segmentation by reinforcing data selection
Jul 24, 2026Chenlan Zhao, Benny Wong, Timothy F. Lundberg +8Few-Shot Medical Image Segmentation Few-Shot Segmentation
Active Learning with Foundation Model Priors: Efficient Learning under Class Imbalance
May 30, 2026Jiancheng Zhang, Meiqing Li, Qi Zhang +1Active Learning Extreme Class Imbalance
One Knob to Rule Them All: A Unified Optimal Transport View of Cold-Start Active Learning
Aug 4, 2026Ning Zhu, Xiaochuan Ma, Juntao Xu +4Active Learning Cold-Start

Jul 24, 2026 · cs.CVJ/K move · Enter open · S save

Active few-shot segmentation by reinforcing data selection

Chenlan Zhao, Benny Wong, Timothy F. Lundberg, Ahmed M. Elsayed+7

Centre for Bioengineering, School of Engineering and Materials Science, Queen Mary University of London, London, United Kingdom · Digital Environment Research Institute, Queen Mary University of London, London, United Kingdom · UCL Hawkes Institute; Department of Medical Physics and Biomedical Engineering, University College London, London, United Kingdom · Department of Biological and Biomedical Sciences, Yale University, Connecticut, USA · Institute of Biomedical Engineering, Department of Engineering Science, University of Oxford, Oxford, United Kingdom

Few-shot learning enables medical image segmentation models to adapt to new tasks using only a small number of labelled examples. However, adaptation performance depends strongly on which examples are selected for the support set. Effective support sets should capture relevant variation within the target domain and be informative for adaptation, with constituent samples providing complementary information. Despite this, existing active data selection approaches largely prioritise samples individually and do not explicitly account for interactions between examples. In this work, we propose a reinforcement learning framework for support-set selection in few-shot medical image segmentation, enabling support sets to be optimised jointly rather than through independent sample scoring. Given a pool of unlabelled candidate images, an agent directly predicts a support set that maximises downstream segmentation performance. Experiments on a cross-institutional pelvic MRI dataset demonstrate improvements over random selection and current state-of-the-art methods. Our findings highlight the importance of support-set complementarity for effective adaptation and demonstrate the potential of reinforcement learning for optimising adaptation sets.

Few-Shot Medical Image Segmentation Few-Shot Segmentation Parameter-Efficient Adaptation

Open paper Ask about it PDF

Abstract

Explore similar work

Active few-shot segmentation by reinforcing data selection

Active Learning with Foundation Model Priors: Efficient Learning under Class Imbalance

One Knob to Rule Them All: A Unified Optimal Transport View of Cold-Start Active Learning