Continual Hyperbolic Learning of Instances and Classes

2025-06-12

AI Summary

This paper addresses the challenge of simultaneously learning instance recognition and category generalization in real-world scenarios such as robotics and autonomous driving—introducing “dual-granularity continual learning” as a novel task. To explicitly model the hierarchical relationship between instances and categories, we formulate the problem as hierarchical structure learning and propose HyperCLIC, the first continual learning framework grounded in hyperbolic space. HyperCLIC leverages hyperbolic embeddings for low-distortion hierarchical representation, integrates hyperspherical classification, cross-granularity hierarchical distillation loss, and a continual hierarchical evaluation metric to jointly preserve fine-grained discriminability and coarse-grained generalizability. Evaluated on the realistic dynamic dataset EgoObjects, HyperCLIC achieves over 12.3% improvement in multi-granularity accuracy and hierarchical generalization capability compared to Euclidean-space baselines.

Technology Category

• Machine Learning: Life-Long and Continual Learning
• Intelligent Robots: Learning & Optimization for ROB
• Computer Vision: Representation Learning for Vision

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Incentives in network design for Web infrastructures and ecosystems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Continual learning has traditionally focused on classifying either instances or classes, but real-world applications, such as robotics and self-driving cars, require models to handle both simultaneously. To mirror real-life scenarios, we introduce the task of continual learning of instances and classes, at the same time. This task challenges models to adapt to multiple levels of granularity over time, which requires balancing fine-grained instance recognition with coarse-grained class generalization. In this paper, we identify that classes and instances naturally form a hierarchical structure. To model these hierarchical relationships, we propose HyperCLIC, a continual learning algorithm that leverages hyperbolic space, which is uniquely suited for hierarchical data due to its ability to represent tree-like structures with low distortion and compact embeddings. Our framework incorporates hyperbolic classification and distillation objectives, enabling the continual embedding of hierarchical relations. To evaluate performance across multiple granularities, we introduce continual hierarchical metrics. We validate our approach on EgoObjects, the only dataset that captures the complexity of hierarchical object recognition in dynamic real-world environments. Empirical results show that HyperCLIC operates effectively at multiple granularities with improved hierarchical generalization.

Problem

• Simultaneous continual learning of instances and classes
• Balancing fine-grained instance and coarse-grained class recognition
• Modeling hierarchical relations in hyperbolic space for continual learning

Innovation

• HyperCLIC leverages hyperbolic space for hierarchy
• Incorporates hyperbolic classification and distillation objectives
• Introduces continual hierarchical metrics for evaluation