Auto-Meta: Automated Gradient Based Meta Learner Search
Abstract
Fully automating machine learning pipelines is one of the key challenges of current artificial intelligence research, since practical machine learning often requires costly and time-consuming human-powered processes such as model design, algorithm development, and hyperparameter tuning. In this paper, we verify that automated architecture search synergizes with the effect of gradient-based meta learning. We adopt the progressive neural architecture search \cite{liu:pnas_google:DBLP:journals/corr/abs-1712-00559} to find optimal architectures for meta-learners. The gradient based meta-learner whose architecture was automatically found achieved state-of-the-art results on the 5-shot 5-way Mini-ImageNet classification problem with accuracy, which is improvement over the result obtained by the first gradient-based meta-learner called MAML \cite{finn:maml:DBLP:conf/icml/FinnAL17}. To our best knowledge, this work is the first successful neural architecture search implementation in the context of meta learning.
Cite
@article{arxiv.1806.06927,
title = {Auto-Meta: Automated Gradient Based Meta Learner Search},
author = {Jaehong Kim and Sangyeul Lee and Sungwan Kim and Moonsu Cha and Jung Kwon Lee and Youngduck Choi and Yongseok Choi and Dong-Yeon Cho and Jiwon Kim},
journal= {arXiv preprint arXiv:1806.06927},
year = {2018}
}
Comments
Presented at NIPS 2018 Workshop on Meta-Learning (MetaLearn 2018)