Research Group in Intelligent Systems
N�ria Maci� » 2008 » December

Two weeks ago, one of the most important researchers in evolutionary computation visited the GRSI. After 25 years of an intense and extensive research on genetic algorithms, our guest, Prof. David E. Goldberg, focuses his interest on a new field, the entrepreneurship, where he displays his creativity to explore which non-technical skills an engineer should develop and how to adapt the training to attain these qualities. This new Prof. Goldberg’s passion leads him to write his latest book, entitled “The Entrepreneurial Engineer”. Some of his ideas are briefly presented in the Prof. Goldberg’s interview by “El Mundo”, in Spanish language.
 
 

 

Today, Núria Macià  and Rubén Nicolàs have set and explained the basis of their work. Please, join me in congratulating both Rubén and Núria as well as their supervisors Elisabet Golobardes and Ester Bernadó and in encouraging them to continue with their work. The following provides a view summary of the defenses.

DEA Ruben

Rubén’s DEA: At 11:00, Rubén started a nice talk on applying different types of learning algorithms, ranging from clustering techniques to CBR systems alike, to the detection of melanoma cancer. The talk was well developed and presented the hard work done on structuring the data and applying machine learning techniques to the problem. 

 

DEA NuriaNúria’s DEA: Thereafter, Núria started with a nice reflection on the need for characterizing our learning problems to fully capture the sources of difficulties and link them to the properties of different learning systems. With no doubt, this is an ancient goal stated in many fields (and further investigated in areas such as optimization) that was sort of overlooked in data classification but that needs to be addressed to advance in the current research of pattern recognition. The talk quickly shifted to the impressing contributions already provided by the thesis. Núria’s glancing back to some important lessons and pointing toward new promising research lines finished with the amazing talk.

 

Wrapping up: Many congrats Rubén and Núria!

Yesterday, Giovanni Pazienza received his PhD degree. The examining committee awarded the thesis and the presentation, entitled Aspects of algorithms and dynamics of cellular paradigms, with an Excellent Cum Laude. They also promised him a brilliant research career for his talent, dynamism, energy, and for many other of his qualities such as… “universality and fractality”.

Many congratulations and we hope you keep being persistent wherever you are!
 
 

On Friday 12, Albert Orriols received his PhD degree. He presented his work entitled New challenges in learning classifier systems: Mining rarities and evolving fuzzy models with a magnificent mise-en-scène which deserved the most sincere congratulations of the examining committee, especially from Prof. Goldberg who remarked that it was one of the best dissertation defenses he had attended.

To be honest, it was not a surprise, you had worked hard for that.

Many congrats!

The International Conference on Pattern Recognition finished last Thursday in Tampa, Florida. 1631 papers by 3556 authors from 47 countries were submitted of which 1006 were accepted and grouped into 74 regular and 16 poster sessions. In one of the latter, we found a work on synthetic data set generation based on a class separability measure, a little participation of the GRSI leaded by Núria Macià , Ester Bernadó-Mansilla, and Albert Orriols-Puig.
The poster session was great, many people from different places (Cuba, Italy, France, Germany, Russia, Spain, and United Kingdom) were interested in the work and provided us with helpful suggestions to improve and enlarge our framework. So, many thanks to all of them.

S+SSPR, a satellite event of the ICPR, took place last week in the hustle and bustle of the University of Central Florida in Orlando.

Dr. Tin Kam Ho did a talk at S+SSPR entitled Data complexity analysis: Linkage between context and solution in classification. She explained that classification accuracy depends on both learner quality and intrinsic data complexity. So, changes in source data and feature transformations are necessary to reduce the data difficulty and to improve learner performance. Among her collaborators, she mentioned the work performed by Ester Bernadó, Albert Orriols, and Núria Macià. Data complexity is a large knowledge area which should be explored in many directions, Dr. Ho suggested some of them for further research, easier to follow than the conference indications.

This picture was taken by Dr. Ludmila Kuncheva, another invited speaker. With some nice pictures and a fresh touch of humor, she presented Linear discriminant classifier (LDC) for streaming data with concept drift. She provided a framework within which theoretical relationship can be sought between the window size and the classification error.
She is also offering a promising lecture in the ICPR under the interesting title Classifier ensembles: Facts, fiction, faults and future.

Prof. Pedro Domingos presented an impressive work summarized by the title Markov logic: A unifying language for structural and statistical pattern recognition. It is a powerful language that combines statistical and structural aspects of the input data. Models in Markov logic are sets of weighted formulas in first-order logic which are interpreted as templates for features of Markov random fields.

Finally, Dr. Horst Bunke presented Graph classification on dissimilarity space embedding. He proposed a methodology that consists in embedding graphs in vector spaces and then applying a statistical classifier to outperform classifiers that directly operate in the graph domain. It is a contribution towards unifying the domains of structural and statistical pattern recognition. We highlight a graph data base repository that contains data sets covering a wide spectrum of different applications.

For further information, please refer to S+SSPR web site.

Today starts the “Research Week”. From Wednesday to Wednesday the campus la Salle offers plenty of research activities originated from the GRSI: lectures performed by world-known researchers and presentations of thesis and DEAs.

Among these activities Prof. Goldberg and Prof. Chua will delight us with their talks. No doubt, this is a rendezvous that an engineer cannot miss!

You can find below the summary of the program schedule:

Wednesday, 10 17:00 Sala de Graus Lecture
      Prof. David E. Goldberg
      The creativity imperative and the technology professional of the future
Thursday, 11 18:00 Auditori Lecture
      Prof. David E. Goldberg
      Not your grandmother’s genetic algorithm
Friday, 12 11:00 Sala de Graus Thesis presentation
      Albert Orriols
      New challenges in learning classifier systems: Mining rarities and evolving fuzzy models
  17:00 Auditori Lecture
      Prof. Leon O. Chua
      Memristor: 37 years later
Monday, 15 11:00 Sala de Graus Thesis presentation
      Giovanni Pazienza
      Aspects of algorithms and dynamics of cellular paradigms
Wednesday, 17 10:30 Sala de Graus DEA presentation
      Rubén Nicolàs
  11:30 Sala de Graus DEA presentation
      Núria Macià
      Anàlisi de la relació entre la complexitat del problema i la precisió dels classificadors en mineria de dades

* All of the activities are taking place in Ed. Sant Josep - Enginyeria i Arquitectura La Salle.
* For further details, please refer to the previous posts.

Prof. Chua

Anyone who ever took an electronics laboratory class will be familiar with the fundamental passive circuit elements: the resistor, the capacitor and the inductor. However, in 1971 Leon Chua reasoned from symmetry arguments that there should be a fourth fundamental element, which he called a memristor (short for memory resistor). Although he showed that such an element has many interesting and valuable circuit properties, until now no one has presented either a useful physical model or an example of a memristor. Here we show, using a simple analytical example, that memristance arises naturally in nanoscale systems in which solid-state electronic and ionic transport are coupled under an external bias voltage.

One application of the memristor could be to create a new type of ultradense non-volatile random access memory (RAM) that would supplement and eventually replace today’s commonly used dynamic random access memory (DRAM). Computers using conventional DRAM lack the ability to retain information once they lose power. When power is restored to a DRAM-based computer, a slow, energy-consuming “boot-up” process is necessary to retrieve data from a magnetic disk required to run the system. In contrast, a memristor-based computer would retain its information after losing power and would not require the boot-up process, resulting in the consumption of less power and wasted time.

Another potential application of memristor technology could be the development of computer systems that remember and associate series of events in a manner similar to the way a human brain recognizes patterns. This could substantially improve today’s facial recognition technology, enable security and privacy features that recognize a complex set of biometric features of an authorized person to access personal information, or enable an appliance to learn from experience.

Prof. Leon O. Chua will extensible talk about memristor in Enginyeria i Arquitectura La Salle on December 12 at 17:00. Go here to sign up for the talk.

Prof. David E. Goldberg will embark us in an interesting journey that starts considering simple GAs and follows the road to the design of competent GA, breaking the billion-variable optimization barrier. The talk will be held in Enginyeria i Arquitectura La Salle on December 11 at 18:00. Go here to sign up. In what follows, further information is provided.

Speaker: David E. Goldberg

Title: Not Your Grandmother’s Genetic Algorithm

Abstract: Genetic algorithms (GAs)–search procedures inspired by the mechanics of natural selection and genetics–have been increasingly applied across the spectrum of human endeavor. Genetic algorithms were also a primary ingredient in the rise in interest in the study of complex adaptive and emergent systems. Despite these salutary effects, some persist in thinking of evolutionary processes–and algorithms that rely upon them–as inherently slow, unreliable, and ad hoc, without much theoretical support. This talk briefly introduces GAs, but quickly shifts to a line of work that has succeeded in supporting GA mechanics with bounding design theory that has been used to demonstrate GA scalability, speed, and range of reliable applicability. Key elements of this theory are discussed to give insight into this accomplishment and to make the point that fast, scalable GAs may also be viewed as first-order models of human innovative or inventive processes. The talk highlights recent results in breaking the billion-variable optimization barrier for the first time. It also discusses recent experience with GA-inspired creativity or innovation support systems.

Bio: David E. Goldberg, a leader in the field of genetic algorithms, is the Jerry S. Dobrovolny Distinguished Professor in Entrepreneurial Engineering at the University of Illinois at Urbana-Champaign. He is also co-founder and chief scientist of ShareThis Inc., a web2.0 startup company. Trained as a civil engineer at the University of Michigan, where he earned his B.S.E. and took his Ph.D. in 1983, Dr. Goldberg has held positions at Michigan, Alabama, and Illinois. He is co-chair of the inaugural and 2nd Workshop on Philosophy and Engineering (TUDelft & the Royal Academy of Engineeirng), and he was recently name co-director the Illinois Foundation for Innovation in Engineering Education. . Among many honors, he is the recipient of a National Science Foundation Presidential Young Investigator Award, a Wickenden Award presented by the American Society for Engineering Education, and an Outstanding Instructor Award presented by the National Technological University . In addition to articles in professional journals, he is the author of two books on genetic algorithms, the widely-cited Genetic Algorithms in Search, Organization, and Machine Learning (1989) and The Design of Innovation (2002), and, most recently, The Entrepreneurial Engineer, which was published in 2006 by Wiley.

Goldberg Prof. David E. Goldberg is giving a talk about the technology professional of the future. The talk will be held in Enginyeria i Arquitectura La Salle on December 10 at 17:00. Go here to sign up. In what follows, the further information about the talk is provided.

Title: The Creativity Imperative and the Technology Professional of the Future

Speaker: David E. Goldberg, Co-Director, Illinois Foundry for Innovation in Engineering Education

Abstract: The world (1) is apparently flat, (2) is being given over to a rising creative class, and (3) requires a whole new mind, but a common conclusion drawn from authors such as Friedman, Florida, and Pink is that technology professionals in advanced economies must excel at creating new categories of product and service, as returns to routine engineering/technology labor are declining because of the ease with which these tasks may be outsourced. This talk starts by examining the setting after World War 2 that has largely shaped engineering and technology education of the recent era. It continues by discussing the techno-economic forces that have affected the intervening time, and it considers recent work by Price in others to understand the essential characteristics and habits of tech visionaries (TVs) who currently help major companies create value through the effective bootstrapping of entirely new product lines. The talk concludes by discussing the newly established Illinois Foundry for Innovation in Engineering Education, including crucial organizational, content, and delivery innovations designed to promote effective and viral curriculum reform.

Bio: David E. Goldberg, a leader in the field of genetic algorithms, is the Jerry S. Dobrovolny Distinguished Professor in Entrepreneurial Engineering at the University of Illinois at Urbana-Champaign. He is also co-founder and chief scientist of ShareThis Inc., a web2.0 startup company. Trained as a civil engineer at the University of Michigan, where he earned his B.S.E. and took his Ph.D. in 1983, Dr. Goldberg has held positions at Michigan, Alabama, and Illinois. He is co-chair of the inaugural and 2nd Workshop on Philosophy and Engineering (TUDelft & the Royal Academy of Engineeirng), and he was recently name co-director the Illinois Foundation for Innovation in Engineering Education. . Among many honors, he is the recipient of a National Science Foundation Presidential Young Investigator Award, a Wickenden Award presented by the American Society for Engineering Education, and an Outstanding Instructor Award presented by the National Technological University . In addition to articles in professional journals, he is the author of two books on genetic algorithms, the widely-cited Genetic Algorithms in Search, Organization, and Machine Learning (1989) and The Design of Innovation (2002), and, most recently, The Entrepreneurial Engineer, which was published in 2006 by Wiley.