Dynamic Sub-graph Distillation for Robust Semi-supervised Continual Learning
This work addresses the challenge of reducing labeled data requirements for continual learning, making it more practical for real-life deployment, though it appears incremental as it builds on existing CL methods.
The paper tackles the problem of semi-supervised continual learning (SSCL), where models learn from partially labeled data with unknown categories, by proposing Dynamic Sub-Graph Distillation (DSGD) to stabilize training against unreliable unlabeled data distributions, achieving improved performance in mitigating catastrophic forgetting across datasets like CIFAR10, CIFAR100, and ImageNet-100 with varying supervision ratios.
Continual learning (CL) has shown promising results and comparable performance to learning at once in a fully supervised manner. However, CL strategies typically require a large number of labeled samples, making their real-life deployment challenging. In this work, we focus on semi-supervised continual learning (SSCL), where the model progressively learns from partially labeled data with unknown categories. We provide a comprehensive analysis of SSCL and demonstrate that unreliable distributions of unlabeled data lead to unstable training and refinement of the progressing stages. This problem severely impacts the performance of SSCL. To address the limitations, we propose a novel approach called Dynamic Sub-Graph Distillation (DSGD) for semi-supervised continual learning, which leverages both semantic and structural information to achieve more stable knowledge distillation on unlabeled data and exhibit robustness against distribution bias. Firstly, we formalize a general model of structural distillation and design a dynamic graph construction for the continual learning progress. Next, we define a structure distillation vector and design a dynamic sub-graph distillation algorithm, which enables end-to-end training and adaptability to scale up tasks. The entire proposed method is adaptable to various CL methods and supervision settings. Finally, experiments conducted on three datasets CIFAR10, CIFAR100, and ImageNet-100, with varying supervision ratios, demonstrate the effectiveness of our proposed approach in mitigating the catastrophic forgetting problem in semi-supervised continual learning scenarios.