Meta-learning agents excel at rapidly learning new tasks from open-ended task distributions; yet, they forget what they learn about each task as soon as the next begins. When tasks reoccur - as they do in natural environments - metalearning agents must explore again instead of immediately exploiting previously discovered solutions. We propose a formalism for generating open-ended yet repetitious environments, then develop a meta-learning architecture for solving these environments. This architecture melds the standard LSTM working memory with a differentiable neural episodic memory. We explore the capabilities of agents with this episodic LSTM in five meta-learning environments with reoccurring tasks, ranging from bandits to navigation and stochastic sequential decision problems.
@article{arxiv.1805.09692,
title = {Been There, Done That: Meta-Learning with Episodic Recall},
author = {Samuel Ritter and Jane X. Wang and Zeb Kurth-Nelson and Siddhant M. Jayakumar and Charles Blundell and Razvan Pascanu and Matthew Botvinick},
journal= {arXiv preprint arXiv:1805.09692},
year = {2018}
}