Unraveling hidden interactions in complex systems with deep learning

Abstract Rich phenomena from complex systems have long intrigued researchers, and yet modeling system micro-dynamics and inferring the forms of interaction remain challenging for conventional data-driven approaches, being generally established by scientists with human ingenuity. In this study, we pr...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Seungwoong Ha, Hawoong Jeong
Formato:	article
Lenguaje:	EN
Publicado:	Nature Portfolio 2021
Materias:	Medicine R Science Q
Acceso en línea:	https://doaj.org/article/9b596c2186b345608377c08d8edfb791
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:9b596c2186b345608377c08d8edfb791
record_format	dspace
spelling	oai:doaj.org-article:9b596c2186b345608377c08d8edfb7912021-12-02T17:39:57ZUnraveling hidden interactions in complex systems with deep learning10.1038/s41598-021-91878-w2045-2322https://doaj.org/article/9b596c2186b345608377c08d8edfb7912021-06-01T00:00:00Zhttps://doi.org/10.1038/s41598-021-91878-whttps://doaj.org/toc/2045-2322Abstract Rich phenomena from complex systems have long intrigued researchers, and yet modeling system micro-dynamics and inferring the forms of interaction remain challenging for conventional data-driven approaches, being generally established by scientists with human ingenuity. In this study, we propose AgentNet, a model-free data-driven framework consisting of deep neural networks to reveal and analyze the hidden interactions in complex systems from observed data alone. AgentNet utilizes a graph attention network with novel variable-wise attention to model the interaction between individual agents, and employs various encoders and decoders that can be selectively applied to any desired system. Our model successfully captured a wide variety of simulated complex systems, namely cellular automata (discrete), the Vicsek model (continuous), and active Ornstein–Uhlenbeck particles (non-Markovian) in which, notably, AgentNet’s visualized attention values coincided with the true variable-wise interaction strengths and exhibited collective behavior that was absent in the training data. A demonstration with empirical data from a flock of birds showed that AgentNet could identify hidden interaction ranges exhibited by real birds, which cannot be detected by conventional velocity correlation analysis. We expect our framework to open a novel path to investigating complex systems and to provide insight into general process-driven modeling.Seungwoong HaHawoong JeongNature PortfolioarticleMedicineRScienceQENScientific Reports, Vol 11, Iss 1, Pp 1-13 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Medicine R Science Q
spellingShingle	Medicine R Science Q Seungwoong Ha Hawoong Jeong Unraveling hidden interactions in complex systems with deep learning
description	Abstract Rich phenomena from complex systems have long intrigued researchers, and yet modeling system micro-dynamics and inferring the forms of interaction remain challenging for conventional data-driven approaches, being generally established by scientists with human ingenuity. In this study, we propose AgentNet, a model-free data-driven framework consisting of deep neural networks to reveal and analyze the hidden interactions in complex systems from observed data alone. AgentNet utilizes a graph attention network with novel variable-wise attention to model the interaction between individual agents, and employs various encoders and decoders that can be selectively applied to any desired system. Our model successfully captured a wide variety of simulated complex systems, namely cellular automata (discrete), the Vicsek model (continuous), and active Ornstein–Uhlenbeck particles (non-Markovian) in which, notably, AgentNet’s visualized attention values coincided with the true variable-wise interaction strengths and exhibited collective behavior that was absent in the training data. A demonstration with empirical data from a flock of birds showed that AgentNet could identify hidden interaction ranges exhibited by real birds, which cannot be detected by conventional velocity correlation analysis. We expect our framework to open a novel path to investigating complex systems and to provide insight into general process-driven modeling.
format	article
author	Seungwoong Ha Hawoong Jeong
author_facet	Seungwoong Ha Hawoong Jeong
author_sort	Seungwoong Ha
title	Unraveling hidden interactions in complex systems with deep learning
title_short	Unraveling hidden interactions in complex systems with deep learning
title_full	Unraveling hidden interactions in complex systems with deep learning
title_fullStr	Unraveling hidden interactions in complex systems with deep learning
title_full_unstemmed	Unraveling hidden interactions in complex systems with deep learning
title_sort	unraveling hidden interactions in complex systems with deep learning
publisher	Nature Portfolio
publishDate	2021
url	https://doaj.org/article/9b596c2186b345608377c08d8edfb791
work_keys_str_mv	AT seungwoongha unravelinghiddeninteractionsincomplexsystemswithdeeplearning AT hawoongjeong unravelinghiddeninteractionsincomplexsystemswithdeeplearning
_version_	1718379778606628864

Unraveling hidden interactions in complex systems with deep learning

Ejemplares similares