Meta-Curvature
This work addresses the challenge of fast adaptation and generalization in few-shot learning for machine learning practitioners, representing an incremental advancement over existing meta-learning approaches.
The authors tackled the problem of few-shot learning by proposing meta-curvature, a framework that learns curvature information to improve generalization and accelerate model adaptation, achieving substantial improvements over previous methods and faster convergence rates without task-specific techniques.
We propose meta-curvature (MC), a framework to learn curvature information for better generalization and fast model adaptation. MC expands on the model-agnostic meta-learner (MAML) by learning to transform the gradients in the inner optimization such that the transformed gradients achieve better generalization performance to a new task. For training large scale neural networks, we decompose the curvature matrix into smaller matrices in a novel scheme where we capture the dependencies of the model's parameters with a series of tensor products. We demonstrate the effects of our proposed method on several few-shot learning tasks and datasets. Without any task specific techniques and architectures, the proposed method achieves substantial improvement upon previous MAML variants and outperforms the recent state-of-the-art methods. Furthermore, we observe faster convergence rates of the meta-training process. Finally, we present an analysis that explains better generalization performance with the meta-trained curvature.