MAS – Agent and Systems Modelling
Luís Correia
The Agent and Systems Modelling Group (MAS) carries out research in the area of Artificial Intelligence in general, with a focus on agent based approaches for modelling intelligent systems. Within BioISI, MAS explores links between biology and physics with artificial intelligence, self-organized systems and bio-inspired computation.
MAS is organised as a whole research group, nevertheless formation of informal sub-groups is encouraged and this has occurred to carry out research focused on two specific thematic areas, namely: GUESS: focusing on social simulation; GruVA: dedicated to artificial life. Research is otherwise organised in a flat structure, easing cooperation among members around different themes.
Specific research topics pursued by MAS include: 1) Artificial Intelligence, with a special focus on agents, including aspects such as agent modelling, agent architectures, multi-agent models and architectures, machine learning, emotion modelling, mobile robotics, self-organisation and social simulation; 2) Agent Societies, including mobile robotics and evolutionary models for building artificial neural networks, swarm cognition, evolutionary algorithms for multi-set representation of populations and operators and symbiogenetic approaches; 3) Data Visualization models of augmented reality and interactive Virtual Reality applications for phobia or anxiety treatments; 4) Data mining and knowledge discovery using large datasets from Omics; 5) Development of tools for integrative visualization of biodata, and user-friendly accessibility to biodata (Biodatabases).
The integration of MAS activities within BioISI develop effortlessly from the bridges that are established between artificial life and biology, as well as the link between data mining and bioinformatics. On the other hand, research in areas like social simulation, agent modelling and visualization are clearly extension areas that enrich and complete the research of BioISI Biology-oriented groups.
Team
PI’s: Luís Correia | Beatriz Carmo | Helder Coelho | Luís Antunes | Paulo Urbano | Luís Cavique | Ana Paula Claúdio | Isabel Nunes | João Pedro Neto | Paulo Trigo | Graça Gaspar | Luís Morgado | Luís Moniz
Recent Publications
Coelho, F., Neto, J.P. (2017). A Method for Regularization of Evolutionary Polynomial Regression, Applied Soft Computing Journal 59C, pp. 223-228.
Gomes, J., Mariano P., Christensen, A.L. (2017). Dynamic Team Heterogeneity in Cooperative Coevolutionary Algorithms. IEEE Transactions on Evolutionary Computation
Silva, F., Correia, L., Christensen, A.L., (2017). Evolutionary online behaviour learning and adaptation in real robots. Royal Society open science , 4:160938