Kaichen Ouyang - Personal Website

About Me ：Kaichen Ouyang

I graduated with a Bachelor's degree from the School of Mathematical Sciences at the University of Science and Technology of China in June 2024. My research interests lie in artificial intelligence, cognitive science, and complex systems.

Google Scholar: View Profile

Education Background

2020.09 – 2024.06

University of Science and Technology of China

Bachelor of Mathematics and Applied Mathematics

IELTS: 7.0/9.0

Certificate: Plan for strengthening basic academic disciplines, Strengthening Foundation Plan in Mathematics

Conference Experience

5th Amorphous Physics and Materials Symposium 2024, Attendee

IEEE Congress on Evolutionary Computation (CEC) 2025, Oral Presentation

School Experience

2022: Mathematical Analysis B1, Teaching Assistant

2024: Mathematical Modeling, Teaching Assistant

Research Experience

2025.5 - Present

Data-Driven Multi-Objective Evolutionary Design of Battery Liquid Cooling Materials, USTC, RA

2024.2 - Present

Graph Neural Networks & Material Science, Songshan Lake Materials Laboratory, RA

2023.9 - 2024.4

Deep Neural Network-Based Control of Quantum Uncertain Systems, USTC, University Innovation Project

2023.5 - 2024.6

Intersection of Non-equilibrium Statistical Physics & Machine Learning, USTC, RA

2021.9 - Present

Evolutionary Algorithms & Machine Learning, Wenzhou University, RA

2021.3 - 2022.9

Non-equilibrium Statistical Physics & Complex Networks, USTC, RA

Reviewer Experience

2024: Swarm and Evolutionary Computation (JCR Q1, IF: 8.2), Reviewer

2025: Knowledge-Based Systems (JCR Q1, IF: 7.2), Reviewer

2025: International Joint Conference on Neural Networks (IJCNN), Reviewer

2025: International Conference on Intelligent Computing (ICIC), Reviewer

2025: AAAI 2026, Reviewer

2025: Computers and Electrical Engineering (JCR Q1, IF: 4.9), Reviewer

2025: International Journal of Computational Intelligence Systems (JCR Q2, IF: 3.116), Reviewer

2025: Information Sciences (JCR Q1, IF: 6.8), Reviewer

Skills

Language skills: Chinese (Native), English (Fluent)

Computer Skills: Microsoft Office 365, Python, MATLAB, MySQL, Java, C/C++, Lammps

Contact

Email: oykc@mail.ustc.edu.cn

Tel: +86 15888787619

Research Interests Overview

Data-Driven Complex Adaptive Systems for Decision-Making

In nature, biological evolution, physical dynamics, and learning behavior form three interconnected adaptive mechanisms: Biological evolution & swarm intelligence drive adaptation through both genetic variation/natural selection and emergent collective behaviors; physical dynamics govern nonequilibrium systems' self-organization toward equilibrium states (e.g., thermodynamic entropy increase and dissipative structures); while learning enables individual behavioral adaptation through experience and interaction with the environment. This includes the reward hypothesis, where behavior is adapted based on the feedback received from actions, allowing individuals to optimize their strategies to achieve long-term goals. Additionally, neural networks model the brain's learning processes by adjusting the connections between neurons, similar to synaptic plasticity, improving decision-making through experience. Together, these mechanisms shape complex adaptive behaviors across scales.

Specifically, these three adaptive mechanisms correspond to three distinct research domains: (1) evolutionary computation simultaneously models both biological evolution (through population-based genetic optimization) and swarm intelligence behaviors (via emergent agent coordination mechanisms), thereby capturing two fundamental layers of biological adaptation. (2) statistical physics formally characterizes self-organization via disorder-to-order transitions mathematically equivalent to physical phase transitions; and (3) machine learning explicitly emulates biological learning through both neural network training grounded in synaptic plasticity principles and reinforcement learning, where behaviors are adapted based on feedback from interactions with the environment, optimizing strategies to maximize long-term rewards. However, contemporary artificial intelligence has predominantly focused on intelligence paradigms centered around human intelligence (e.e., neural networks simulating cortical information processing), while systematically neglecting two equally fundamental adaptive mechanisms in nature: (1) evolutionary and swarm intelligence in biological systems, and (2) self-organizing intelligence in physical systems. I contend that advancing artificial intelligence requires moving beyond this anthropocentric paradigm to incorporate these more universal forms of natural intelligence. These systems demonstrate unique elegance through their inherent adaptive mechanisms. We are perpetually inspired by nature's ingenuity manifested in such self-organizing intelligence. I maintain that natural adaptation consistently surpasses artificial intervention in efficacy. My primary research therefore focuses on constructing data-driven frameworks to understand and harness these systems for enhanced decision-making.

Research Interests:

Artificial Intelligence Cognitive Science Complex Systems

Publications and Preprints

K Ouyang, S Fu, Y Chen, Q Cai, AA Heidari, H Chen. Escape: an optimization method based on crowd evacuation behaviors. Artificial Intelligence Review 58(1), 2024. First Author https://doi.org/10.1007/s10462-024-11008-6

K Ouyang, D Wei, X Sha, J Yu, Y Zhao, M Qiu, S Fu, AA Heidar, H Chen. Beaver Behavior Optimizer: A Novel Metaheuristic Algorithm for Solar PV Parameter Identification and Engineering Problems. Journal of Advanced Research. First Author https://doi.org/10.1016/j.jare.2025.09.001

K Ouyang, D Wei, S Fu, S Gu, X Sha, J Yu, J Yu, AA Heidar, Z Cai, H Chen. Multi-objective Red-billed Blue Magpie Optimizer: A Novel Algorithm for Multi-objective UAV Path Planning. Results in Engineering. First Author https://doi.org/10.1016/j.rineng.2025.106785

K Ouyang, S Fu, Y Chen, H Chen. Dynamic Graph Neural Evolution: An Evolutionary Framework Integrating Graph Neural Networks with Adaptive Filtering. 2025 IEEE Congress on Evolutionary Computation (Oral). First Author https://ieeexplore.ieee.org/document/11042917

D Wei, Z Wang, M Qiu, J Yu, J Yu, Y Jin, X Sha, K Ouyang. Multiple Objectives Escaping Bird Search Optimization and Its application in Stock Market Prediction Based on Transformer Model. Scientific Reports 15(1), 5730, 2025. Corresponding Author https://doi.org/10.1038/s41598-025-88883-8

K Ouyang, et al. A Comprehensive Analysis of Digital Inclusive Finance's Influence on High Quality Enterprise Development through Fixed Effects and Deep Learning Frameworks. Scientific Reports. Corresponding Author https://doi.org/10.1038/s41598-025-14610-y

S Qiu, Y Wang, Z Ke, Q Shen, Z Li, R Zhang, K Ouyang. A Generative Adversarial Network-Based Investor Sentiment Indicator. Mathematics 13(9), 1476, 2025. Corresponding Author https://doi.org/10.3390/math13091476

JB Lian, K Ouyang, R Zhong, Y Zhang, S Luo, L Ma, X Wu, H Chen. Trend-Aware Mechanism for metaheuristic algorithms. Applied Soft Computing, 2025. Second Author https://doi.org/10.1016/j.asoc.2025.113505

YQ Wang, C Xu, ML Fang, TZ Li, LW Zhang, DS Wei, K Ouyang, et al. Study of nonequilibrium phase transitions mechanisms in exclusive network and node model of heterogeneous assignment based on real experimental data of KIF3AC and KIF3CC motors. The European Physical Journal Plus 137(10), 1-22, 2022. Co Author https://doi.org/10.1140/epjp/s13360-022-03372-5

YQ Wang, DS Wei, LW Zhang, TY Zhang, TZ Li, ML Fang, KC Ouyang, et al. Physical mechanisms of exit dynamics in microchannels. International Journal of Modern Physics B 38(15), 2450193. Co Author https://doi.org/10.1142/S0217979224501935

K Ouyang‡, S Zhang‡, S Liu‡, J Tian, Y Li, H Tong, H Bai, YC Hu, WH Wang. Graph Learning Metallic Glass Discovery from Wikipedia. Nature Machine Intelligence (Under Review). First Author https://doi.org/10.48550/arXiv.2507.19536

K Ouyang, S Fu. Learn from Global Correlations: Enhancing Evolutionary Algorithm via Spectral GNN. arXiv preprint arXiv:2412.17629, 2024. First Author https://doi.org/10.48550/arXiv.2412.17629

S Gu, ..., K Ouyang, et al. Wave Optics Optimizer: A novel meta-heuristic algorithm for engineering optimization. Communications In Nonlinear Science And Numerical Simulation. Co Author https://www.sciencedirect.com/science/article/pii/S1007570425007464

K Ouyang, T Hou, JJ Lian, S Fu, Z Ke, D Wei, M Qiu, J Ouyang. Stochastic Gradient-guided Adaptive Differential Evolution: Algorithm and Its Application in the Diagnosis of COVID-19, Influenza, and Bacterial Pneumonia. Artificial Intelligence In Medicine (Under Review). First Author

K Ouyang. Rethinking Over-Smoothing in Graph Neural Networks: A Perspective from Anderson Localization. arXiv preprint arXiv:2507.05263, 2025. Sole First Author https://doi.org/10.48550/arXiv.2507.05263

K Ouyang. Consciousness as a Jamming Phase. arXiv preprint arXiv:2507.08197, 2025. Sole First Author https://arxiv.org/abs/2507.08197

K Ouyang. Why Flow Matching is Particle Swarm Optimization?. arXiv preprint arXiv:2507.20810, 2025. Sole First Author https://arxiv.org/abs/2507.20810

J Yu, J Yu, D Wei, X Sha, S Fu, M Qiu, Y Jin, K Ouyang. Multi-Objective Mobile Damped Wave Algorithm (MOMDWA): A Novel Approach For Quantum System Control. arXiv preprint arXiv:2502.05228, 2025. Corresponding Author https://arxiv.org/abs/2502.05228

W Xiao, JJ Lian, K Ouyang, S Gu, Z Ke, D Wei, X Sha, J Wang, S Fu, M Qiu, C Xu. Newton Downhill Optimizer for Global Optimization with Application to Breast Cancer Feature Selection. Biomedical Signal Processing and Control (Under Review). Corresponding Author

D Wei, K Ouyang, Z Wang, X Sha, M Qiu, Z Yi, AA Heidari, H Chen. Multi-strategy boosted dung beetle algorithm and its application for bankruptcy prediction. Neural Networks (Under Review). Corresponding Author

JJ Lian, H Chen, K Ouyang, et al. Twisted Convolutional Networks (TCNs): Enhancing Feature Interactions for Non-Spatial Data Classification. Neural Networks (Under Review). Co Author

JJ Lian, K Ouyang, et al. IKUN: A mean-field game theoretic KD-tree density guided mechanism for swarm optimization. Swarm and Evolutionary Computation (Under Review). Co Author