Mitigating Label Noise using Prompt-Based Hyperbolic Meta-Learning in Open-Set Domain Generalization

2024-12-24

AI Summary

This work addresses the practical challenge of label noise in open-set domain generalization (OSDG), proposing the first noisy-label OSDG benchmark (OSDG-NL). To jointly mitigate label noise and domain shift, we introduce HyProMeta—a novel framework that constructs class prototypes in hyperbolic space to enhance discriminability and incorporates learnable, class-agnostic prompts for noise-aware meta-learning. By integrating hyperbolic geometric modeling, prompt-based learning, and label denoising strategies, HyProMeta significantly improves both known-class classification accuracy and unknown-class rejection capability. Extensive experiments on noisy variants of PACS and DigitsDG demonstrate that HyProMeta consistently outperforms existing state-of-the-art methods across all evaluation metrics. The code is publicly available.

Technology Category

• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Search and Optimization: Metareasoning and Metaheuristics
• Computer Vision: Adversarial Attacks & Robustness

Application Category

• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Open-Set Domain Generalization (OSDG) is a challenging task requiring models to accurately predict familiar categories while minimizing confidence for unknown categories to effectively reject them in unseen domains. While the OSDG field has seen considerable advancements, the impact of label noise--a common issue in real-world datasets--has been largely overlooked. Label noise can mislead model optimization, thereby exacerbating the challenges of open-set recognition in novel domains. In this study, we take the first step towards addressing Open-Set Domain Generalization under Noisy Labels (OSDG-NL) by constructing dedicated benchmarks derived from widely used OSDG datasets, including PACS and DigitsDG. We evaluate baseline approaches by integrating techniques from both label denoising and OSDG methodologies, highlighting the limitations of existing strategies in handling label noise effectively. To address these limitations, we propose HyProMeta, a novel framework that integrates hyperbolic category prototypes for label noise-aware meta-learning alongside a learnable new-category agnostic prompt designed to enhance generalization to unseen classes. Our extensive experiments demonstrate the superior performance of HyProMeta compared to state-of-the-art methods across the newly established benchmarks. The source code of this work is released at https://github.com/KPeng9510/HyProMeta.

Problem

• Open Set Domain Generalization
• Label Noise
• Recognition Accuracy

Innovation

• HyProMeta
• Hyperbolic Geometry
• Noisy Labels OSDG