Decoupling Adaptation from Modeling with Meta-Optimizers for Meta Learning

by   Sébastien M. R. Arnold, et al.

Meta-learning methods, most notably Model-Agnostic Meta-Learning or MAML, have achieved great success in adapting to new tasks quickly, after having been trained on similar tasks. The mechanism behind their success, however, is poorly understood. We begin this work with an experimental analysis of MAML, finding that deep models are crucial for its success, even given sets of simple tasks where a linear model would suffice on any individual task. Furthermore, on image-recognition tasks, we find that the early layers of MAML-trained models learn task-invariant features, while later layers are used for adaptation, providing further evidence that these models require greater capacity than is strictly necessary for their individual tasks. Following our findings, we propose a method which enables better use of model capacity at inference time by separating the adaptation aspect of meta-learning into parameters that are only used for adaptation but are not part of the forward model. We find that our approach enables more effective meta-learning in smaller models, which are suitably sized for the individual tasks.


page 1

page 2

page 3

page 4


The effects of negative adaptation in Model-Agnostic Meta-Learning

The capacity of meta-learning algorithms to quickly adapt to a variety o...

Meta-Learning of Neural State-Space Models Using Data From Similar Systems

Deep neural state-space models (SSMs) provide a powerful tool for modeli...

Learning to Learn and Sample BRDFs

We propose a method to accelerate the joint process of physically acquir...

Model-Agnostic Meta-Learning using Runge-Kutta Methods

Meta-learning has emerged as an important framework for learning new tas...

Meta-learning within Projective Simulation

Learning models of artificial intelligence can nowadays perform very wel...

Meta-learning of Pooling Layers for Character Recognition

In convolutional neural network-based character recognition, pooling lay...

Fast and Effective Adaptation of Facial Action Unit Detection Deep Model

Detecting facial action units (AU) is one of the fundamental steps in au...

Code Repositories


Code release for "When MAML Can Adapt Fast and How to Assist When It Cannot", AISTATS 2021.

view repo

Please sign up or login with your details

Forgot password? Click here to reset