Deep Meta-Learning: Learning to Learn in the Concept Space
About
Few-shot learning remains challenging for meta-learning that learns a learning algorithm (meta-learner) from many related tasks. In this work, we argue that this is due to the lack of a good representation for meta-learning, and propose deep meta-learning to integrate the representation power of deep learning into meta-learning. The framework is composed of three modules, a concept generator, a meta-learner, and a concept discriminator, which are learned jointly. The concept generator, e.g. a deep residual net, extracts a representation for each instance that captures its high-level concept, on which the meta-learner performs few-shot learning, and the concept discriminator recognizes the concepts. By learning to learn in the concept space rather than in the complicated instance space, deep meta-learning can substantially improve vanilla meta-learning, which is demonstrated on various few-shot image recognition problems. For example, on 5-way-1-shot image recognition on CIFAR-100 and CUB-200, it improves Matching Nets from 50.53% and 56.53% to 58.18% and 63.47%, improves MAML from 49.28% and 50.45% to 56.65% and 64.63%, and improves Meta-SGD from 53.83% and 53.34% to 61.62% and 66.95%, respectively.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Few-shot Image Classification | Mini-Imagenet (test) | Accuracy71.3 | 235 | |
| 5-way Classification | miniImageNet (test) | -- | 231 | |
| Few-shot classification | CUB (test) | Accuracy77.1 | 145 | |
| 5-way Few-shot Classification | CUB | 5-shot Acc77.1 | 95 | |
| 5-way Few-shot Classification | CIFAR100 | 1-shot Acc61.6 | 6 | |
| 5-way Few-shot Classification | Caltech-256 | 1-shot Accuracy62.2 | 6 |