Modeling Collective Behavior for Fish School With Deep Q-Networks

Modeling collective behavior is a way to better understand the mechanisms that govern collective animal behaviors. Traditional rule-based modeling methods rely heavily on human prior knowledge and may not provide a proper explanation of the phenomenon of collective behaviors. This paper proposes a D...

Full description

Saved in:
Bibliographic Details
Main Authors: Pengyu Chen, Fang Wang, Shuo Liu, Yifan Yu, Shengzhi Yue, Yanan Song, Yuanshan Lin
Format: Article
Language:English
Published: IEEE 2023-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10087222/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849334749377593344
author Pengyu Chen
Fang Wang
Shuo Liu
Yifan Yu
Shengzhi Yue
Yanan Song
Yuanshan Lin
author_facet Pengyu Chen
Fang Wang
Shuo Liu
Yifan Yu
Shengzhi Yue
Yanan Song
Yuanshan Lin
author_sort Pengyu Chen
collection DOAJ
description Modeling collective behavior is a way to better understand the mechanisms that govern collective animal behaviors. Traditional rule-based modeling methods rely heavily on human prior knowledge and may not provide a proper explanation of the phenomenon of collective behaviors. This paper proposes a Deep Q-Networks (DQN)-based modeling method for fish school. Firstly, an individual’s state (continuous value) is expressed by the angle between its direction and the average direction of its perceived neighbors. An individual’s action is represented with discretized turning angle. Then, the reward function is constructed with the change in the number of neighbors. And finally, the neural network structure is constructed to represent the Q-value function and is trained by the DQN algorithm. The proposed approach is tested in two scenarios: single-learner and multi-learner. Results show that in both scenarios the proposed method can gradually converge and finally obtain a model that can produce collective behavior. On this basis, this paper also deeply analyzes the learned model from the perspectives of average order parameter and collective behavior patterns. It verifies that the behavior pattern generated by the learned model is a highly ordered collective behavior. In addition, we make a comparison between our proposed approach and the Q-Learning algorithm. The results show that our approach not only has a stronger ability to express policy and is better at handling continuous states but also has a more stable learning performance in training.
format Article
id doaj-art-402aff8758434d8dbab8444281c3c3bb
institution Kabale University
issn 2169-3536
language English
publishDate 2023-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-402aff8758434d8dbab8444281c3c3bb2025-08-20T03:45:28ZengIEEEIEEE Access2169-35362023-01-0111366303664110.1109/ACCESS.2023.326323710087222Modeling Collective Behavior for Fish School With Deep Q-NetworksPengyu Chen0https://orcid.org/0000-0002-9323-8936Fang Wang1Shuo Liu2Yifan Yu3Shengzhi Yue4Yanan Song5Yuanshan Lin6https://orcid.org/0000-0001-9053-8004College of Information Engineering, Dalian Ocean University, Dalian, ChinaCollege of Information Engineering, Dalian Ocean University, Dalian, ChinaCollege of Information Engineering, Dalian Ocean University, Dalian, ChinaCollege of Information Engineering, Dalian Ocean University, Dalian, ChinaCollege of Information Engineering, Dalian Ocean University, Dalian, ChinaCollege of Information Engineering, Dalian Ocean University, Dalian, ChinaCollege of Information Engineering, Dalian Ocean University, Dalian, ChinaModeling collective behavior is a way to better understand the mechanisms that govern collective animal behaviors. Traditional rule-based modeling methods rely heavily on human prior knowledge and may not provide a proper explanation of the phenomenon of collective behaviors. This paper proposes a Deep Q-Networks (DQN)-based modeling method for fish school. Firstly, an individual’s state (continuous value) is expressed by the angle between its direction and the average direction of its perceived neighbors. An individual’s action is represented with discretized turning angle. Then, the reward function is constructed with the change in the number of neighbors. And finally, the neural network structure is constructed to represent the Q-value function and is trained by the DQN algorithm. The proposed approach is tested in two scenarios: single-learner and multi-learner. Results show that in both scenarios the proposed method can gradually converge and finally obtain a model that can produce collective behavior. On this basis, this paper also deeply analyzes the learned model from the perspectives of average order parameter and collective behavior patterns. It verifies that the behavior pattern generated by the learned model is a highly ordered collective behavior. In addition, we make a comparison between our proposed approach and the Q-Learning algorithm. The results show that our approach not only has a stronger ability to express policy and is better at handling continuous states but also has a more stable learning performance in training.https://ieeexplore.ieee.org/document/10087222/Collective behaviorcollective behavior modelDeep Q-Networks (DQN)fish school
spellingShingle Pengyu Chen
Fang Wang
Shuo Liu
Yifan Yu
Shengzhi Yue
Yanan Song
Yuanshan Lin
Modeling Collective Behavior for Fish School With Deep Q-Networks
IEEE Access
Collective behavior
collective behavior model
Deep Q-Networks (DQN)
fish school
title Modeling Collective Behavior for Fish School With Deep Q-Networks
title_full Modeling Collective Behavior for Fish School With Deep Q-Networks
title_fullStr Modeling Collective Behavior for Fish School With Deep Q-Networks
title_full_unstemmed Modeling Collective Behavior for Fish School With Deep Q-Networks
title_short Modeling Collective Behavior for Fish School With Deep Q-Networks
title_sort modeling collective behavior for fish school with deep q networks
topic Collective behavior
collective behavior model
Deep Q-Networks (DQN)
fish school
url https://ieeexplore.ieee.org/document/10087222/
work_keys_str_mv AT pengyuchen modelingcollectivebehaviorforfishschoolwithdeepqnetworks
AT fangwang modelingcollectivebehaviorforfishschoolwithdeepqnetworks
AT shuoliu modelingcollectivebehaviorforfishschoolwithdeepqnetworks
AT yifanyu modelingcollectivebehaviorforfishschoolwithdeepqnetworks
AT shengzhiyue modelingcollectivebehaviorforfishschoolwithdeepqnetworks
AT yanansong modelingcollectivebehaviorforfishschoolwithdeepqnetworks
AT yuanshanlin modelingcollectivebehaviorforfishschoolwithdeepqnetworks