Luís Rocha

FLAD and the Portuguese Scientists in America

Luis Mateus Rocha was born in Luanda, on the 5th of October 1966.
His research and academic career has been developed in the interdisciplinary area of systems science and complex systems, and more specifically in the sub-fields of complex adaptive systems (CAS) and computational intelligence (CI).
His main interest is informational properties of natural and artificial systems, which enable them to adapt and evolve. Since there are many adaptive and evolving systems in nature and technology, this kind of research is necessarily interdisciplinary, and the onus of sound interdisciplinary research, crucially, is the ability to cross fields without trivializing them.

This is particularly important when one uses ideas from, or applies methods to, biology since it is such an empirical and specialized field. Because of this, Luís Rocha has invested much of his research time and effort in the understanding of biology while aiming at interdisciplinary knowledge transfer and with this has contributed with novel insights from biology and cognition to CAS and CI, and, conversely, has also contributed with computational and mathematical methods developed in those fields to tackle problems in biology and cognitive science.

Currently Luis Rocha positions are as an Associate Professor of Informatics, core faculty of the Cognitive Science Program, Adjunct Associate Professor in Computer Science, and affiliated with the Biocomplexity Institute at Indiana University, Bloomington, USA. He exerts also the cargo of director of the Computational Biology Collaboratorium and the direction of the PhD program in Computational Biology at the Instituto Gulbenkian da Ciencia, Portugal.

His formal education begun with a Licenciatura in Mechanical Engineering at Instituto Superior Tecnico in Lisbon, during 1985 through 1990, and in his last year (1990) he was also an Erasmus Exchange Student of Industrial Automation in North Staffordshire University. After graduation, he had a one-year position as a Graduate Research Assistant at Laboratório Nacional de Engenharia Civil (1990-1991) and in 1992 he went to the USA to start his PhD in Systems Science at the State University of New York at Binghamton, which he completed in 1997.

During 1997-1998 he was a Postdoc at Los Alamos National Laboratory and after that, between 1999 and 2004 he achieved a position as a Permanent Staff Scientist and Team Leader at that institution.
Since 2004 that he has been an Associate Professor of School of Informatics and Cognitive Science Program at Indiana University.

During his undergraduate education on Mechanical Engineering at the Instituto Superior Técnico in Lisbon, he became specialized on systems engineering and industrial automation (1985-1990). More specifically, he focused on Artificial Intelligence and Robotics. After a semester spent at Staffordshire University in England, working on Industrial Automation he became very interested in adaptive technology. Thus, upon his return, he started working with Pedro Medina-Martins on cybenetics, evolutionary systems, fuzzy set theory and generally on cybernetics and systems thinking.

During the development of his undergraduate thesis he learned more about systems science and began working on extending some aspects of Gordon Pask’s conversation theory using fuzzy set theory. In particular, he developed a database for academic libraries, which used Hebbian learning and fuzzy information retrieval to adaptively recognize the interests of users. Interestingly, Gordon Pask had developed a computer system (Thoughtsticker) in the 1970's which greatly resembled the modern day World Wide Web, but which additionally provided mechanisms to organize links according to patterns of use.

In early 1991 a lot of excitement was generated in systems science and cybernetics circles about Berners-Lee’s World Wide Web project (whose first web page and browser had just been published in late 1990). Given the obvious similarities between the newly developed WWW and Pask’s work, Luís Rocha was encouraged to submit his undergraduate thesis as a paper to the first workshop of the Principa Cybernetica Project organized at the Free University of Brussels, where Gordon Pask and several other cyberneticians would be present.

Gordon Pask became very interested in Luís Rocha’s work and encouraged him to pursue a PhD, introducing him to Heinz von Foerster, one of the original members of the Cybernetic Group at the Macy Conferences in the early 1950's. Luís Rocha then started a correspondence with both of these influential systems scientists that lasted until their deaths (Pask in 1996, and von Foerster in 2002). The work of these two scientists has been crucial for his early research development and he wrote several papers continuing their work in the 1990's. Given his interest in fuzzy set theory and models of uncertainty, both of them introduced him and encouraged him to pursue his PhD with George Klir, one of the most respected mathematical systems theorists and author of highly influential books on systems science, information and uncertainty theory.

Thus, he enrolled in the PhD program in Systems Science at the State University of New York in Binghamton in 1992, one of the only programs of its kind in the World. He applied for a loan from a Portuguese bank (with a 20% interest rate!) to obtain the necessary funds for a single semester in New York, hoping to secure a grant later on. Indeed, empowered with a strong reference from Fernando Carvalho Rodrigues, he applied and was granted a PhD Scholarship from the old Junta Nacional de Investigação Científica (Programa Ciência), and this grant gave him the security to persevere his PhD studies in a focused manner.

Instead of having a single PhD advisor, he opted to work with two co-advisors: the mathematician George Klir, and the theoretical biologist Howard Pattee. In addition, he took various courses in the Cognitive Science program at SUNY. His doctoral education was thus inherently interdisciplinary but still focused in the area of complex systems, which lead him to spend some time at the Santa Fe Institute in the mid-nineties. From the beginning, his work has been centered on computational approaches to complex systems, with applications to biology, web technology and cognitive science.

He has published 50 peer-reviewed articles in many of the top journals and conferences in systems science and complex adaptive systems. He has also edited four peer-reviewed volumes including the Artificial Life X proceedings published by MIT Press and special issues of Biosystems and Artificial Life.

After his PhD from the State University of New York, he has only worked in two institutions: The Los Alamos National Laboratory and Indiana University. He moved from Los Alamos to Indiana for various reasons, but mostly because of the academic freedom provided by Indiana University, which had been declining at the Los Alamos National Laboratory under the current climate of security concerns. Moreover, the School of Informatics at Indiana University is one of the leaders of the new, so-called i-schools focusing on interdisciplinary aspects of information technology. Finally, but very important for the decision to move there Luís Rocha says, “the Cognitive Science program at IU, besides being one of the most respected in the World, has been the key center of the approach to cognitive science I identify myself with.”

His research has been supported in the U.S. by federal grants or projects (DOE, NSF, Homeland Security, etc). He has raised several million dollars as PI or co-PI in these grants. In Portugal, he has participated in the fund raising efforts to establish Computational Biology initiatives at the Instituto Gulbenkian de Ciência. These endeavors include the PhD program on Computational Biology and the FLAD Computational Biology Collaboratorium.

In a long-term perspective his goals are to better understand the nature of complex networks of information exchange in biology and technology, and to use the knowledge of such networks to produce informatics technology to integrate extremely large collections of data, and also to predict trends and behaviors from incomplete network information.

“As for a long-term dream, I'd like to contribute to our knowledge of the origin of information from a purely dynamical system---ultimately, to contribute a little bit towards our understanding of the origin of life”, says Luís Rocha.

The scientific community mostly knows Luís Rocha due to his work on adaptive recommendation systems at Los Alamos. He has designed biologically-inspired algorithms that allow databases of documents and the web to adapt to the changing interests of the communities of users of such systems. Another work for which he is recognized is his model of the evolutionary advantages of RNA Editing, which grants phenotypic plasticity to organisms. “RNA Editing could grant organisms, computational power to "re-program" their phenotypes according to environmental clues”, Luís Rocha proposed this in the mid-1990s and presently modern evidence seems to confirm that claim.

His work has been covered in the media, including MSNBC, the Washington Post, O Estado de São Paulo, Expresso, Techpoint, Radio-Televisão Portuguesa, The Santa Fe New Mexican, The Indianapolis Star, and Lusa News Agency. He has also contributed small opinion pieces that were published in the New York Times Book Review and Wired Magazine among others. He has received three best paper awards in the leading Systems Science conference: the European Meeting on Cybernetics and Systems Research (EMCSR).

He has given invited or plenary talks at more than 40 venues such as universities, research institutes, conferences and workshops. Examples of these venues are Stanford University, University of Michigan, Oxford University, University of Kyoto, University of Illinois, NASA Goddard Space Center, Xerox PARC, the Konrad Lorenz Institute, the Gordon Conference on RNA Editing, the Joint Statistical Meetings, several workshops at the Santa Fe Institute, etc.

Concerning Portugal, Luís Rocha says he has received a couple of compelling offers to return, but in general, the host universities did not have strong graduate programs, which would make it very difficult to pursue his research. On other occasions, when he had shown interest in applying to officially opened positions, he was discouraged from applying because the positions were really already assigned to someone "in house".

However, Luís Rocha never lost touch with the Portuguese scientific community. Through the years, he has helped the Fundação para a Ciência e Tecnologia with grant reviewing, he has been a visiting professor at the Instituto de Sistemas e Robótica at the Instituto Superior Técnico, and at the Instituto Gulbenkian de Ciência, where currently he directs the FLAD Computational Biology Collaboratorium with a three month per year commitment. Luís Rocha also adds, “I am always open to returning to Portugal more permanently if the right conditions are met.”

rocha at indiana.edu

Comentários