A Machine With Human-Like Memory Systems

Taewoon Kim; Michael Cochez; Vincent Francois Lavet; Mark A. Neerincx; Piek Vossen

A Machine With Human-Like Memory Systems

Taewoon Kim, Michael Cochez, Vincent Francois Lavet, Mark A. Neerincx, Piek Vossen

Research output: Chapter in Book / Report / Conference proceeding › Conference contribution › Academic › peer-review

Abstract

Inspired by the cognitive science theory, we explicitly model an agent with both semantic and episodic memory systems, and show that it is better than having just one of the two memory systems. In order to show this, we have designed and released our own challenging environment, "the Room", compatible with OpenAI Gym, where an agent has to properly learn how to encode, store, and retrieve memories to maximize its rewards. The Room environment allows for a hybrid intelligence setup where machines and humans can collaborate. We show that two agents collaborating with each other results in better performance than one agent acting alone. We have open-sourced our code and models at this https URL.

Original language	English
Title of host publication	Proceedings of the First international workshop on Human-Centered Design of Symbiotic Hybrid Intelligence
Publisher	HHAI
Number of pages	7
Publication status	E-pub ahead of print - 4 Apr 2022

Access to Document

Persistent URL (handle)

Copy link address

Cite this

@inproceedings{319bd3c34e364297b06ae8d69793f94f,

title = "A Machine With Human-Like Memory Systems",

abstract = "Inspired by the cognitive science theory, we explicitly model an agent with both semantic and episodic memory systems, and show that it is better than having just one of the two memory systems. In order to show this, we have designed and released our own challenging environment, {"}the Room{"}, compatible with OpenAI Gym, where an agent has to properly learn how to encode, store, and retrieve memories to maximize its rewards. The Room environment allows for a hybrid intelligence setup where machines and humans can collaborate. We show that two agents collaborating with each other results in better performance than one agent acting alone. We have open-sourced our code and models at this https URL.",

author = "Taewoon Kim and Michael Cochez and {Francois Lavet}, Vincent and Neerincx, {Mark A.} and Piek Vossen",

year = "2022",

month = apr,

day = "4",

language = "English",

booktitle = "Proceedings of the First international workshop on Human-Centered Design of Symbiotic Hybrid Intelligence",

publisher = "HHAI",

}

TY - GEN

T1 - A Machine With Human-Like Memory Systems

AU - Kim, Taewoon

AU - Cochez, Michael

AU - Francois Lavet, Vincent

AU - Neerincx, Mark A.

AU - Vossen, Piek

PY - 2022/4/4

Y1 - 2022/4/4

N2 - Inspired by the cognitive science theory, we explicitly model an agent with both semantic and episodic memory systems, and show that it is better than having just one of the two memory systems. In order to show this, we have designed and released our own challenging environment, "the Room", compatible with OpenAI Gym, where an agent has to properly learn how to encode, store, and retrieve memories to maximize its rewards. The Room environment allows for a hybrid intelligence setup where machines and humans can collaborate. We show that two agents collaborating with each other results in better performance than one agent acting alone. We have open-sourced our code and models at this https URL.

AB - Inspired by the cognitive science theory, we explicitly model an agent with both semantic and episodic memory systems, and show that it is better than having just one of the two memory systems. In order to show this, we have designed and released our own challenging environment, "the Room", compatible with OpenAI Gym, where an agent has to properly learn how to encode, store, and retrieve memories to maximize its rewards. The Room environment allows for a hybrid intelligence setup where machines and humans can collaborate. We show that two agents collaborating with each other results in better performance than one agent acting alone. We have open-sourced our code and models at this https URL.

UR - https://ii.tudelft.nl/humancenteredsymbioticHI/

M3 - Conference contribution

BT - Proceedings of the First international workshop on Human-Centered Design of Symbiotic Hybrid Intelligence

PB - HHAI

ER -

A Machine With Human-Like Memory Systems

Abstract

Access to Document

Persistent URL (handle)

Other files and links

Fingerprint

Cite this