MALADY: Multiclass Active Learning with Auction Dynamics on Graphs

2024-09-14

arXiv.org

AI Summary

This work addresses active learning for multi-class semi-supervised classification. We propose the first active learning framework grounded in auction dynamics on graphs. Our method constructs a sample similarity graph and formulates classification as generalized energy functional optimization; crucially, it innovatively leverages dual variables of the auction algorithm to quantify boundary uncertainty, enabling efficient and interpretable sample querying. This constitutes the first systematic extension of auction dynamics to multi-class active learning, integrating graph neural networks with dual optimization theory. Evaluated on multiple standard benchmarks, our approach achieves an average accuracy improvement of 3.2–5.8% over state-of-the-art methods under identical labeling budgets, demonstrating substantial gains in labeling efficiency—particularly in low-label regimes—and enhanced model generalization.

Technology Category

• Machine Learning: Active Learning
• Search and Optimization: Learning to Search
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Advertising auctions, pricing, markets, and exchanges

Abstract

Active learning enhances the performance of machine learning methods, particularly in semi-supervised cases, by judiciously selecting a limited number of unlabeled data points for labeling, with the goal of improving the performance of an underlying classifier. In this work, we introduce the Multiclass Active Learning with Auction Dynamics on Graphs (MALADY) framework which leverages the auction dynamics algorithm on similarity graphs for efficient active learning. In particular, we generalize the auction dynamics algorithm on similarity graphs for semi-supervised learning in [24] to incorporate a more general optimization functional. Moreover, we introduce a novel active learning acquisition function that uses the dual variable of the auction algorithm to measure the uncertainty in the classifier to prioritize queries near the decision boundaries between different classes. Lastly, using experiments on classification tasks, we evaluate the performance of our proposed method and show that it exceeds that of comparison algorithms.

Problem

• Develops active learning for multiclass classification on graphs
• Generalizes auction dynamics for semi-supervised learning optimization
• Proposes acquisition function using dual variables for boundary queries

Innovation

• Auction dynamics algorithm on similarity graphs
• Generalized optimization functional for multiclass learning
• Dual variable-based active learning acquisition function