My primary research interests lie at the intersection of two fields: evolutionary computation and multi-agent systems. I am specifically interested in how evolutionary search algorithms can be used to help people understand and analyze agent-based models of complex systems (e.g., flocking birds, traffic jams, or how information diffuses across social networks). My secondary research interests broadly span the areas of artificial life, multi-agent robotics, cognitive/learning science, design of multi-agent modeling environments. I enjoy interdisciplinary research, and in pursuit of the aforementioned topics, I have been involved in application areas from archeology to zoology, from linguistics to marketing, and from urban growth patterns to materials science. I am also very interested in creative approaches to computer science and complex systems education, and have published work on the use of multi-agent simulation as a vehicle for introducing students to computer science.
It is my philosophy that theoretical research should be inspired by real-world problems, and conversely, that theoretical results should inform and enhance practice in the field. Accordingly, I view tool building as a vital practice that is complementary to theoretical and methodological research. Throughout my own work I have contributed to the research community by developing several practical software tools, including BehaviorSearch (http://www.behaviorsearch.org/)
Kenneth D. Aiello is a postdoctoral research scholar with the Global BioSocial Complexity Initiative at ASU. Kenneth’s research contributes to cross disciplinary conversations on how historical developments in biological, social, and cultural knowledge systems are governed by processes that transform the structure, dynamics, and function of complex systems. Applying computational historical analysis and epistemology to question what scientific knowledge is and how we can analyze changes in knowledge, he uses text analysis, social network analysis, and machine learning to measure similarities and differences between the knowledge claims of individual agents and groups. His work builds on how to assess contested knowledge claims and measure the evolution of knowledge across complex systems and multiple dimensions of scale. This approach also engages in dynamic new debates about global and local structures of knowledge shaped by technological innovation within microbiology related to public policy, shrinking resources given to biomedical ideas as opposed to “translation”, and the ethics of scientific discovery. Using interdisciplinary methods for understanding historical content and context rich narratives contributes to understanding new domains and major transitions in science and provides a richer understanding of how knowledge emerges.
Furkan Gürsoy received the BS in Management Information Systems from Boğaziçi University, Turkey, and the MS in Data Science from İstanbul Şehir University, Turkey. He is currently a PhD Candidate at Boğaziçi University. He previously worked as an IS/IT Consultant and a Machine Learning Engineer with the industry for several years. He held a Visiting Researcher Position with IMT Atlantique, France, in 2020. His research interests include complex networks, machine learning, simulation, and broad data science.
network science, machine learning, simulation, data science.
coupling of agent-based models and mathematical models
machine learning algorithms
deep learning algorithms
infectious diseases modeling
Mainly interested in studying social networks of learners, teachers, and innovators. Uses Social Network Analysis, but also sentiment analysis, data mining, and recommender system techniques.
I am investigating the use of machine learning techniques in non-stationary modeling environments to better reproduce aspects of human learning and decision-making in human-natural system simulations.
Dr. Mariam Kiran is a Research Scientist at LBNL, with roles at ESnet and Computational Research Division. Her current research focuses on deep reinforcement learning techniques and multi-agent applications to optimize control of system architectures such as HPC grids, high-speed networks and Cloud infrastructures.. Her work involves optimization of QoS, performance using parallelization algorithms and software engineering principles to solve complex data intensive problems such as large-scale complex decision-making. Over the years, she has been working with biologists, economists, social scientists, building tools and performing optimization of architectures for multiple problems in their domain.