Toward Design of Synthetic Active Inference Agents by Mere Mortals

A. (Bert) de Vries

doi:10.1007/978-3-031-47958-8_11

Toward Design of Synthetic Active Inference Agents by Mere Mortals

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

Samenvatting

The theoretical properties of active inference agents are impressive, but how do we realize effective agents in working hardware and software on edge devices? This is an interesting problem because the computational load for policy exploration explodes exponentially, while the computational resources are very limited for edge devices. In this paper, we discuss the necessary features for a software toolbox that supports a competent non-expert engineer to develop working active inference agents. We introduce a toolbox-in-progress that aims to accelerate the democratization of active inference agents in a similar way as TensorFlow propelled applications of deep learning technology.

Originele taal-2	Engels
Titel	Active Inference
Subtitel	4th International Workshop, IWAI 2023, Ghent, Belgium, September 13–15, 2023, Revised Selected Papers
Redacteuren	Christopher L. Buckley, Daniela Cialfi, Pablo Lanillos, Maxwell Ramstead, Noor Sajid, Hideaki Shimazaki, Tim Verbelen, Martijn Wisse
Plaats van productie	Cham
Uitgeverij	Springer
Pagina's	173-185
Aantal pagina's	13
ISBN van elektronische versie	978-3-031-47958-8
ISBN van geprinte versie	978-3-031-47957-1
DOI's	https://doi.org/10.1007/978-3-031-47958-8_11
Status	Gepubliceerd - 2024
Evenement	4th International Workshop on Active Inference, IWAI 2023 - Ghent, België Duur: 13 sep. 2023 → 15 sep. 2023

Publicatie series

Naam	Communications in Computer and Information Science (CCIS)
Volume	1915
ISSN van geprinte versie	1865-0929
ISSN van elektronische versie	1865-0937

Congres

Congres	4th International Workshop on Active Inference, IWAI 2023
Verkorte titel	IWAI 2023
Land/Regio	België
Stad	Ghent
Periode	13/09/23 → 15/09/23

Toegang tot document

10.1007/978-3-031-47958-8_11

Andere bestanden en links

Link naar publicatie in Scopus

Citeer dit

de Vries, A. (2024). Toward Design of Synthetic Active Inference Agents by Mere Mortals. In C. L. Buckley, D. Cialfi, P. Lanillos, M. Ramstead, N. Sajid, H. Shimazaki, T. Verbelen, & M. Wisse (editors), Active Inference: 4th International Workshop, IWAI 2023, Ghent, Belgium, September 13–15, 2023, Revised Selected Papers (blz. 173-185). (Communications in Computer and Information Science (CCIS); Vol. 1915). Springer. https://doi.org/10.1007/978-3-031-47958-8_11

de Vries, A. (Bert). / Toward Design of Synthetic Active Inference Agents by Mere Mortals. Active Inference: 4th International Workshop, IWAI 2023, Ghent, Belgium, September 13–15, 2023, Revised Selected Papers. redacteur / Christopher L. Buckley ; Daniela Cialfi ; Pablo Lanillos ; Maxwell Ramstead ; Noor Sajid ; Hideaki Shimazaki ; Tim Verbelen ; Martijn Wisse. Cham : Springer, 2024. blz. 173-185 (Communications in Computer and Information Science (CCIS)).

@inproceedings{9c536dbb6b96458f9531f865b184faa5,

title = "Toward Design of Synthetic Active Inference Agents by Mere Mortals",

abstract = "The theoretical properties of active inference agents are impressive, but how do we realize effective agents in working hardware and software on edge devices? This is an interesting problem because the computational load for policy exploration explodes exponentially, while the computational resources are very limited for edge devices. In this paper, we discuss the necessary features for a software toolbox that supports a competent non-expert engineer to develop working active inference agents. We introduce a toolbox-in-progress that aims to accelerate the democratization of active inference agents in a similar way as TensorFlow propelled applications of deep learning technology.",

keywords = "active inference agent, factor graphs, free energy minimization, reactive message passing, rxinfer, structural adaptation",

author = "{de Vries}, {A. (Bert)}",

year = "2024",

doi = "10.1007/978-3-031-47958-8_11",

language = "English",

isbn = "978-3-031-47957-1",

series = "Communications in Computer and Information Science (CCIS)",

publisher = "Springer",

pages = "173--185",

editor = "Buckley, {Christopher L.} and Daniela Cialfi and Pablo Lanillos and Maxwell Ramstead and Noor Sajid and Hideaki Shimazaki and Tim Verbelen and Martijn Wisse",

booktitle = "Active Inference",

address = "Germany",

note = "4th International Workshop on Active Inference, IWAI 2023, IWAI 2023 ; Conference date: 13-09-2023 Through 15-09-2023",

}

de Vries, A 2024, Toward Design of Synthetic Active Inference Agents by Mere Mortals. in CL Buckley, D Cialfi, P Lanillos, M Ramstead, N Sajid, H Shimazaki, T Verbelen & M Wisse (uitgave), Active Inference: 4th International Workshop, IWAI 2023, Ghent, Belgium, September 13–15, 2023, Revised Selected Papers. Communications in Computer and Information Science (CCIS), vol. 1915, Springer, Cham, blz. 173-185, 4th International Workshop on Active Inference, IWAI 2023, Ghent, België, 13/09/23. https://doi.org/10.1007/978-3-031-47958-8_11

Toward Design of Synthetic Active Inference Agents by Mere Mortals. / de Vries, A. (Bert).
Active Inference: 4th International Workshop, IWAI 2023, Ghent, Belgium, September 13–15, 2023, Revised Selected Papers. uitgave / Christopher L. Buckley; Daniela Cialfi; Pablo Lanillos; Maxwell Ramstead; Noor Sajid; Hideaki Shimazaki; Tim Verbelen; Martijn Wisse. Cham: Springer, 2024. blz. 173-185 (Communications in Computer and Information Science (CCIS); Vol. 1915).

Onderzoeksoutput: Hoofdstuk in Boek/Rapport/Congresprocedure › Conferentiebijdrage › Academic › peer review

TY - GEN

T1 - Toward Design of Synthetic Active Inference Agents by Mere Mortals

AU - de Vries, A. (Bert)

PY - 2024

Y1 - 2024

N2 - The theoretical properties of active inference agents are impressive, but how do we realize effective agents in working hardware and software on edge devices? This is an interesting problem because the computational load for policy exploration explodes exponentially, while the computational resources are very limited for edge devices. In this paper, we discuss the necessary features for a software toolbox that supports a competent non-expert engineer to develop working active inference agents. We introduce a toolbox-in-progress that aims to accelerate the democratization of active inference agents in a similar way as TensorFlow propelled applications of deep learning technology.

AB - The theoretical properties of active inference agents are impressive, but how do we realize effective agents in working hardware and software on edge devices? This is an interesting problem because the computational load for policy exploration explodes exponentially, while the computational resources are very limited for edge devices. In this paper, we discuss the necessary features for a software toolbox that supports a competent non-expert engineer to develop working active inference agents. We introduce a toolbox-in-progress that aims to accelerate the democratization of active inference agents in a similar way as TensorFlow propelled applications of deep learning technology.

KW - active inference agent

KW - factor graphs

KW - free energy minimization

KW - reactive message passing

KW - rxinfer

KW - structural adaptation

UR - http://www.scopus.com/inward/record.url?scp=85177868163&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-47958-8_11

DO - 10.1007/978-3-031-47958-8_11

M3 - Conference contribution

SN - 978-3-031-47957-1

T3 - Communications in Computer and Information Science (CCIS)

SP - 173

EP - 185

BT - Active Inference

A2 - Buckley, Christopher L.

A2 - Cialfi, Daniela

A2 - Lanillos, Pablo

A2 - Ramstead, Maxwell

A2 - Sajid, Noor

A2 - Shimazaki, Hideaki

A2 - Verbelen, Tim

A2 - Wisse, Martijn

PB - Springer

CY - Cham

T2 - 4th International Workshop on Active Inference, IWAI 2023

Y2 - 13 September 2023 through 15 September 2023

ER -

de Vries A. Toward Design of Synthetic Active Inference Agents by Mere Mortals. In Buckley CL, Cialfi D, Lanillos P, Ramstead M, Sajid N, Shimazaki H, Verbelen T, Wisse M, redacteurs, Active Inference: 4th International Workshop, IWAI 2023, Ghent, Belgium, September 13–15, 2023, Revised Selected Papers. Cham: Springer. 2024. blz. 173-185. (Communications in Computer and Information Science (CCIS)). doi: 10.1007/978-3-031-47958-8_11

Toward Design of Synthetic Active Inference Agents by Mere Mortals

Samenvatting

Publicatie series

Congres

Toegang tot document

Andere bestanden en links

Vingerafdruk

Citeer dit