Is Meta-Learning Out? Rethinking Unsupervised Few-Shot Classification with Limited Entropy

2025-09-16

AI Summary

This paper addresses the challenge of fair comparison between meta-learning and full-class supervised training in unsupervised few-shot classification. To enable principled evaluation, we propose an entropy-constrained assessment framework that reveals meta-learning’s distinct advantages under low-entropy regimes, label noise, and task heterogeneity. We introduce MINO—a novel unsupervised meta-learning framework—featuring (i) a dynamic-head DBSCAN module for adaptive construction of unsupervised meta-tasks, and (ii) a stability-weighted meta-scaler to enhance robustness against label noise. Furthermore, we integrate entropy regularization with theoretical analysis to ensure interpretability and generalization guarantees. Extensive experiments on multiple unsupervised few-shot and zero-shot benchmarks demonstrate that MINO consistently outperforms state-of-the-art methods, especially under high label noise and task distribution shift. Our work establishes a new paradigm for applying meta-learning in realistic weakly supervised settings and provides strong empirical validation.

Technology Category

Machine Learning: Semi-Supervised LearningSearch and Optimization: Metareasoning and MetaheuristicsReasoning under Uncertainty: Stochastic Optimization

Application Category

Web Mining and Content Analysis: Robustness and generalizability of Web mining methodsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingGraph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Meta-learning is a powerful paradigm for tackling few-shot tasks. However, recent studies indicate that models trained with the whole-class training strategy can achieve comparable performance to those trained with meta-learning in few-shot classification tasks. To demonstrate the value of meta-learning, we establish an entropy-limited supervised setting for fair comparisons. Through both theoretical analysis and experimental validation, we establish that meta-learning has a tighter generalization bound compared to whole-class training. We unravel that meta-learning is more efficient with limited entropy and is more robust to label noise and heterogeneous tasks, making it well-suited for unsupervised tasks. Based on these insights, We propose MINO, a meta-learning framework designed to enhance unsupervised performance. MINO utilizes the adaptive clustering algorithm DBSCAN with a dynamic head for unsupervised task construction and a stability-based meta-scaler for robustness against label noise. Extensive experiments confirm its effectiveness in multiple unsupervised few-shot and zero-shot tasks.

Problem

• Evaluating meta-learning versus whole-class training for few-shot classification
• Establishing entropy-limited supervised setting for fair comparison
• Proposing robust meta-learning framework for unsupervised tasks

Innovation

• Meta-learning with limited entropy
• DBSCAN adaptive clustering algorithm
• Stability-based meta-scaler robustness