Efficient Automatic Meta Optimization Search for Few-Shot Learning
This work addresses the need for efficient and automated meta-learning methods in computer vision, though it is incremental as it applies existing NAS techniques to meta-learning.
The paper tackles the problem of automating meta-learning architecture design for few-shot learning by using neural architecture search, achieving competitive results on Mini-ImageNet and Omniglot with high transferability and reduced search time to 1-2 GPU days.
Previous works on meta-learning either relied on elaborately hand-designed network structures or adopted specialized learning rules to a particular domain. We propose a universal framework to optimize the meta-learning process automatically by adopting neural architecture search technique (NAS). NAS automatically generates and evaluates meta-learner's architecture for few-shot learning problems, while the meta-learner uses meta-learning algorithm to optimize its parameters based on the distribution of learning tasks. Parameter sharing and experience replay are adopted to accelerate the architectures searching process, so it takes only 1-2 GPU days to find good architectures. Extensive experiments on Mini-ImageNet and Omniglot show that our algorithm excels in few-shot learning tasks. The best architecture found on Mini-ImageNet achieves competitive results when transferred to Omniglot, which shows the high transferability of architectures among different computer vision problems.