Title
A Machine with Short-Term, Episodic, and Semantic Memory Systems
Author
Kim, Taewoon (Vrije Universiteit Amsterdam)
Cochez, Michael (Vrije Universiteit Amsterdam)
François-Lavet, Vincent (Vrije Universiteit Amsterdam)
Neerincx, M.A. (TU Delft Interactive Intelligence) 
Vossen, Piek (Vrije Universiteit Amsterdam)
Contributor
Williams, Brian (editor)
Chen, Yiling (editor)
Neville, Jennifer (editor)
Date
2023
Abstract
Inspired by the cognitive science theory of the explicit human memory systems, we have modeled an agent with short-term, episodic, and semantic memory systems, each of which is modeled with a knowledge graph. To evaluate this system and analyze the behavior of this agent, we designed and released our own reinforcement learning agent environment, “the Room”, where an agent has to learn how to encode, store, and retrieve memories to maximize its return by answering questions. We show that our deep Q-learning based agent successfully learns whether a short-term memory should be forgotten, or rather be stored in the episodic or semantic memory systems. Our experiments indicate that an agent with human-like memory systems can outperform an agent without this memory structure in the environment.
To reference this document use:
http://resolver.tudelft.nl/uuid:d1374e6c-27cb-407e-82a8-b617974a525b
DOI
https://doi.org/10.1609/aaai.v37i1.25075
Publisher
American Association for Artificial Intelligence (AAAI)
Embargo date
2023-12-27
ISBN
978-157735880-0
Source
AAAI-23 Technical Tracks 1
Event
37th AAAI Conference on Artificial Intelligence, AAAI 2023, 2023-02-07 → 2023-02-14, Washington, United States
Series
Proceedings of the 37th AAAI Conference on Artificial Intelligence, AAAI 2023, 37 (1)
Bibliographical note
Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.
Part of collection
Institutional Repository
Document type
conference paper
Rights
© 2023 Taewoon Kim, Michael Cochez, Vincent François-Lavet, M.A. Neerincx, Piek Vossen