Connectionism and Cognitive Science Lab
Welcome to the new L3C website! Here you can find information about our current researchs and publications
News
May 14, 2012 · The L3C group has grown
Check out the researchers of the L3C.
February 22, 2012 · Paper accepted at IEEE CEC 2012
The following paper was accepted for oral presentation at IEEE Congress on Evolutionary Computation that will be presented in Brisbane at June 14th 2012: Cassio Rodrigo Conti, Mauro Roisenberg and Guenther Schwedersky Neto, ACOR-V - an Algorithm that Incorporates the Visibility Heuristic to the ACO in Continuous Domain
February 15, 2011 · L3C website
The L3C website is on line.
Register now to recieve special information about our lab!
Metaheuristic
Term by definition
Methods that optimize a problem by iteratively trying to improve a candidate solution, given a measure of quality. They make few or no assumptions about the problem and can search large spaces of solutions.
Ant Colony Optimization in Continuous Domains
Ant Colony Optimization (ACO) is a meta heuristic based on the foraging behavior of ants. It was initially proposed to solve discrete combinatorial optimization problems. In this project we study some ACO techniques developed to continuous domains problems, comparing the features of classical discrete ACO to the continuous extension. We proposed some own extensions that we believe being more close to the original idea of ACO and that returned better results comparing with others researches.
Esquemas de Aprendizado Neural Aplicados em Robótica Baseada em Comportamento
The degree of autonomy of an agent is related to the ability to decide for itself how to relate the sensor data with the commands to the actuators in their efforts to achieve the purposes for which it was designed. Thus, the ability of learning and adaptation of the agent is closely related to their...
E&PRiskIV
During the development of Module Neural E&PRiskIII, it was observed that a point "critical" in a good performance Module Neural regarding the ability to forecast total operating time with respect to quality of historical data that are used for training neural network and that are in a ...
Officialy founded at may 5th of 2000, this lab combines researchers of Computer Sciences at various fields such as: Artificial Neural Networks, Evolutionary Computation, Game Theory and Logic. Aimed at studying the cognition emergence by Nature inspired paradigms, starting by connectionism making a bridge between symbolic, connectionist and evolutive approaches of Artificial Intelligence. More than a workplace, the lab intends to be a group of researchers linked by common interests without physical frontiers.