An inference perspective on meta-learning

While meta-learning algorithms are often viewed as algorithms that learn to learn, an alternative viewpoint frames meta-learning as inferring a hidden task variable from experience consisting of observations and rewards. From this perspective, learning to learn is learning to infer. This viewpoint can be useful in solving problems in meta-RL, which I’ll demonstrate through two examples: (1) enabling off-policy meta-learning, and (2) performing efficient meta-RL from image observations. I’ll also discuss how this perspective leads to an algorithm for few-shot image segmentation.

An inference perspective on meta-learning

Watch the seminar

Event Information

Abstract

Topics

About the Speaker

Related Seminars

Cause & Consequences of neuronal Tau protein ‘activation’

OpenNeuro FitLins GLM: An Accessible, Semi-Automated Pipeline for OpenNeuro Task fMRI Analysis

The Systems Vision Science Summer School & Symposium, August 11 – 22, 2025, Tuebingen, Germany