Overcoming Catastrophic Forgetting in Incremental Few-Shot Learning by Finding Flat Minima

Abstract

This paper considers incremental few-shot learning, which requires a model to continually recognize new categories with only a few examples provided. Our study shows that existing methods severely suffer from catastrophic forgetting, a well-known problem in incremental learning, which is aggravated due to data scarcity and imbalance in the few-shot setting. Our analysis further suggests that to prevent catastrophic forgetting, actions need to be taken in the primitive stagethe training of base classes instead of later few-shot learning sessions. Therefore, we propose to search for flat local minima of the base training objective function and then fine-tune the model parameters within the flat region on new tasks. In this way, the model can efficiently learn new classes while preserving the old ones. Comprehensive experimental results demonstrate that our approach outperforms all prior state-of-the-art methods and is very close to the approximate upper bound. The source code is available at https://github.com/moukamisama/F2M.

Advancing autonomy through lifelong learning: a survey of autonomous intelligent systems

“…However, as the data-driven algorithms rely on the type, scale, and quality of training data, the coherence, generality, and adaptability of the algorithms across different tasks and environments are great challenges. The challenge for AIS concerned in this paper is the ability to remember previous tasks when learning new ones, known as catastrophic forgetting (Shi et al, 2021). Catastrophic forgetting refers to the phenomenon where a neural network loses previously learned information after training on subsequent tasks, resulting in a drastic performance drop on previous tasks (Serra et al, 2018).…”