The Evolutionary Computation Group at CINVESTAV-IPN (EVOCINV) was formally launched in the year 2001 under the leadership of Dr. Carlos A. Coello Coello. However, the initial members of the group have been working with Dr. Coello since 1999 in different research projects, and therefore the cooperative efforts among several of the group members dates back to that year. The current group consists of 4 full-time PhD students and 2 full-time MSc students who are at CINVESTAV-IPN. There are also 2 more PhD students who do not physically attend CINVESTAV-IPN (one in Sinaloa, Mexico and the other one in San Luis, Argentina), 3 more MSc students from the Universidad Veracruzana (in Xalapa, Veracruz, Mexico), and 2 more BSc students from the Universidad Autónoma de Yucatán (in Mérida, Yucatán). We also collaborate with researchers from CIMAT (in Guanajuato, Mexico), IMTA (in Cuernavaca, Mexico), MTU (in Michigan, USA), the University of Málaga (in Málaga, Spain), Tulane University (in New Orleans, USA), the University of New South Wales (in Australia), the Universidad Autónoma de Yucatán (in Mérida, Mexico), and IIMAS-UNAM (in Mexico City, Mexico).

Our research currently focuses on 3 main areas:

1. Evolutionary Multiobjective Optimization: We have developed a micro-genetic algorithm for multiobjective optimization and an extension of the particle swarm optimization algorithm for multiple objectives. Our current work involves the development and implementation of schemes for the incorporation of user's preferences into evolutionary multiobjective optimization techniques. We are also developing new algorithms and metrics.

2. Constraint-Handling in Evolutionary Algorithms: We have developed 3 new approaches based on multiobjective optimization concepts. Our current work involves the study of the viability of Pareto dominance-based approaches in the context of constraint-handling. We are also developing new constraint-handling schemes based on cultural algorithms and exploring the use of an artificial immune system to handle constraints (both sequentially and in parallel).

3. Evolvable Hardware: We have developed several algorithms for the optimization of combinational circuits at the gate-level, as well as some work on the use of PAC-learning techniques to estimate exact population sizes for a genetic algorithm and the use of genetic programming for designing multiplexers. We are currently exploring the use of multiobjective optimization techniques to reduce the computational requirements of an evolutionary algorithm used for extrinsic evolvable hardware at the gate-level.

Our research is currently sponsored by CONACyT and CINVESTAV-IPN. We currently have several exciting research projects some of which may be on interest for you. Please check our website for further details about our current research and for any research opportunities. We are also always interested in collaborations with other research groups working in related areas, so please do not hesitate to contact us in case you are interested in getting in touch with us.

We welcome you all to our website and we also appreciate your interest in our work.

Sincerely,

Dr. Carlos A. Coello Coello

For further information about EVOCINV, please contact:

Dr. Carlos A. Coello Coello

CINVESTAV-IPN

Sección de Computación

Depto. de Ingeniería Eléctrica

Av. Instituto Politécnico Nacional No. 2508

Col San Pedro Zacatenco

Deleg. Gustavo A. Madero

México, D.F. 07300

MEXICO

email: ccoello@cs.cinvestav.mx

Phone: + 52 55 5061 3800x6564

Fax: + 52 55 5061 3757