Saved in:
Bibliographic Details
Main Authors: Kim, Taewoon, Cochez, Michael, François-Lavet, Vincent, Neerincx, Mark, Vossen, Piek
Format: Preprint
Published: 2022
Subjects:
Online Access:https://arxiv.org/abs/2212.02098
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913135183527936
author Kim, Taewoon
Cochez, Michael
François-Lavet, Vincent
Neerincx, Mark
Vossen, Piek
author_facet Kim, Taewoon
Cochez, Michael
François-Lavet, Vincent
Neerincx, Mark
Vossen, Piek
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2212_02098
institution arXiv
publishDate 2022
record_format arxiv
spellingShingle A Machine with Short-Term, Episodic, and Semantic Memory Systems
Kim, Taewoon
Cochez, Michael
François-Lavet, Vincent
Neerincx, Mark
Vossen, Piek
Artificial Intelligence
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.
title A Machine with Short-Term, Episodic, and Semantic Memory Systems
topic Artificial Intelligence
url https://arxiv.org/abs/2212.02098