Poster

Optimistic Meta-Gradients

Sebastian Flennerhag · Tom Zahavy · Brendan O'Donoghue · Hado van Hasselt · András György · Satinder Singh

Great Hall & Hall B1+B2 (level 1) #1110
[ ]
Wed 13 Dec 3 p.m. PST — 5 p.m. PST

Abstract:

We study the connection between gradient-based meta-learning and convex optimisation. We observe that gradient descent with momentum is a special case of meta-gradients, and building on recent results in optimisation, we prove convergence rates for meta learning in the single task setting. While a meta-learned update rule can yield faster convergence up to constant factor, it is not sufficient for acceleration. Instead, some form of optimism is required. We show that optimism in meta-learning can be captured through the recently proposed Bootstrapped Meta-Gradient (Flennerhag et. al., 2022) method, providing deeper insight into its underlying mechanics.

Chat is not available.