English

A Bridge Between Hyperparameter Optimization and Learning-to-learn

Machine Learning 2019-08-22 v3 Machine Learning

Abstract

We consider a class of a nested optimization problems involving inner and outer objectives. We observe that by taking into explicit account the optimization dynamics for the inner objective it is possible to derive a general framework that unifies gradient-based hyperparameter optimization and meta-learning (or learning-to-learn). Depending on the specific setting, the variables of the outer objective take either the meaning of hyperparameters in a supervised learning problem or parameters of a meta-learner. We show that some recently proposed methods in the latter setting can be instantiated in our framework and tackled with the same gradient-based algorithms. Finally, we discuss possible design patterns for learning-to-learn and present encouraging preliminary experiments for few-shot learning.

Keywords

Cite

@article{arxiv.1712.06283,
  title  = {A Bridge Between Hyperparameter Optimization and Learning-to-learn},
  author = {Luca Franceschi and Michele Donini and Paolo Frasconi and Massimiliano Pontil},
  journal= {arXiv preprint arXiv:1712.06283},
  year   = {2019}
}

Comments

NIPS 2017 workshop on Meta-learning (http://metalearning.ml/)