The Emergence of Abstract and Episodic Neurons in Episodic Meta-RL
Abstract
In this work, we analyze the reinstatement mechanism introduced by Ritter et al. (2018) to reveal two classes of neurons that emerge in the agent's working memory (an epLSTM cell) when trained using episodic meta-RL on an episodic variant of the Harlow visual fixation task. Specifically, Abstract neurons encode knowledge shared across tasks, while Episodic neurons carry information relevant for a specific episode's task.
Cite
Text
AlKhamissi et al. "The Emergence of Abstract and Episodic Neurons in Episodic Meta-RL." ICLR 2021 Workshops: Learning_to_Learn, 2021.Markdown
[AlKhamissi et al. "The Emergence of Abstract and Episodic Neurons in Episodic Meta-RL." ICLR 2021 Workshops: Learning_to_Learn, 2021.](https://mlanthology.org/iclrw/2021/alkhamissi2021iclrw-emergence/)BibTeX
@inproceedings{alkhamissi2021iclrw-emergence,
title = {{The Emergence of Abstract and Episodic Neurons in Episodic Meta-RL}},
author = {AlKhamissi, Badr and ElNokrashy, Muhammad and Spranger, Michael},
booktitle = {ICLR 2021 Workshops: Learning_to_Learn},
year = {2021},
url = {https://mlanthology.org/iclrw/2021/alkhamissi2021iclrw-emergence/}
}