Hypernetwork Approach to Bayesian MAML (Student Abstract)

The main goal of Few-Shot learning algorithms is to enable learning from small amounts of data. One of the most popular and elegant Few-Shot learning approaches is Model-Agnostic Meta-Learning (MAML). In this paper, we propose a novel framework for Bayesian MAML called BH-MAML, which employs Hypernetworks for weight updates. It learns the universal weights point-wise, but a probabilistic structure is added when adapted for specific tasks. In such a framework, we can use simple Gaussian distributions or more complicated posteriors induced by Continuous Normalizing Flows.

Authors

• Piotr Borycki Jagiellonian University
• Piotr Kubacki Jagiellonian University
• Marcin Przewięźlikowski Jagiellonian University
• Tomasz Kuśmierczyk Jagiellonian University
• Jacek Tabor Jagiellonian University
• Przemysław Spurek Jagiellonian University

Keywords

No keywords are indexed for this paper.