Generalized Adaptation for Few-Shot Learning
This work addresses few-shot learning for AI systems by providing an incremental improvement in adaptation methods.
The paper tackles the problem of few-shot learning by proposing a closed-form base learner that constrains adaptation to improve generalization, achieving state-of-the-art performance with 87.75% accuracy on 5-shot miniImageNet, outperforming existing methods by approximately 10%.
Many Few-Shot Learning research works have two stages: pre-training base model and adapting to novel model. In this paper, we propose to use closed-form base learner, which constrains the adapting stage with pre-trained base model to get better generalized novel model. Following theoretical analysis proves its rationality as well as indication of how to train a well-generalized base model. We then conduct experiments on four benchmarks and achieve state-of-the-art performance in all cases. Notably, we achieve the accuracy of 87.75% on 5-shot miniImageNet which approximately outperforms existing methods by 10%.