The LCS and GBML community stop
The third issue of volume 10 of Genetic Programming and Evolvable Machines is now available online, containing the following articles:A three-step decomposition method for the evolutionary design of sequential logic circuitsby Houjun Liang, Wenjian Luo…
The third issue of volume 10 of Genetic Programming and Evolvable Machines is now available online, containing the following articles:
There were several awards presented at the GECCO-2009 conference aside from the Human-Competitive Results Awards (Humies) awards about which Wolfgang posted previously, and for which I’ve listed the other winners below. Of particular interest to reader…
I am pleased to share with you that the Journal of Artificial Evolution and Applications has recently published my LCS Review paper entitled, “Learning Classifier Systems: A Complete Introduction, Review, and Roadmap”. I wrote this from the perspective of a non-computer scientist, to introduce the basic LCS concept, as well as the variation represented in different LCS implementations that have been tasked to different problem domains. It was my goal and hope that this review might provide a reasonable starting point for outsiders interested in understanding or getting involved in the LCS community. This paper may be viewed using the following link: Thanks! I enjoyed listening to the many excellent GBML talks given at GECCO this year.
Below you may find the slides of the GECCO 2009 tutorial that Jaume Bacardit and I put together. Hope you enjoy it.
Slides
Abstract
We are living in the peta-byte era.We have larger and larger data to analyze, process and transform into useful answers for the domain experts. Robust data mining tools, able to cope with petascale volumes […]
Related posts:
Below you may find the slides of the GECCO 2009 tutorial that Jaume Bacardit and I put together. Hope you enjoy it.
Slides
Abstract
We are living in the peta-byte era.We have larger and larger data to analyze, process and transform into useful answers for the domain experts. Robust data mining tools, able to cope with petascale volumes and/or high dimensionality producing human-understandable solutions are key on several domain areas. Genetics-based machine learning (GBML) techniques are perfect candidates for this task, among others, due to the recent advances in representations, learning paradigms, and theoretical modeling. If evolutionary learning techniques aspire to be a relevant player in this context, they need to have the capacity of processing these vast amounts of data and they need to process this data within reasonable time. Moreover, massive computation cycles are getting cheaper and cheaper every day, allowing researchers to have access to unprecedented parallelization degrees. Several topics are interlaced in these two requirements: (1) having the proper learning paradigms and knowledge representations, (2) understanding them and knowing when are they suitable for the problem at hand, (3) using efficiency enhancement techniques, and (4) transforming and visualizing the produced solutions to give back as much insight as possible to the domain experts are few of them.
This tutorial will try to answer this question, following a roadmap that starts with the questions of what large means, and why large is a challenge for GBML methods. Afterwards, we will discuss different facets in which we can overcome this challenge: Efficiency enhancement techniques, representations able to cope with large dimensionality spaces, scalability of learning paradigms. We will also review a topic interlaced with all of them: how can we model the scalability of the components of our GBML systems to better engineer them to get the best performance out of them for large datasets. The roadmap continues with examples of real applications of GBML systems and finishes with an analysis of further directions.
Related posts:
GP was well featured at this year’s GECCO Humies Awards. The most spectacular application which was subsequently awarded first prize (GOLD) was based on two papers by Weimer/Nguyen/Le Goues/Forrestpublished in proceedings of the 31st International Conf…
GP was well featured at this year’s GECCO Humies Awards. The most spectacular application which was subsequently awarded first prize (GOLD) was based on two papers by Weimer/Nguyen/Le Goues/Forrest
published in proceedings of the 31st International Conference on Software Engineering (ICSE) in May 2009 and Forrest/Weimer/Nguyen/Le Goes in this year’s GECCO proceedings. Both papers won awards from the respective conferences, and winning the Humies award was the “icing on the cake”.
The authors apply a specialized/improved form of Genetic Programming to locate and repair software bugs. Repairing software bugs is a time consuming and commercially very costly activity. To date, automating the process has been very difficult. The GP method proposed by our Gold Medal winners takes down the average repair time for software bugs from more than 3 hours per bug to 3 minutes.
The authors rightly claim that “showing how to use GP in the context of modern software systems and integrating GP into modern software practice will help evolutionary computation to become more widely accepted by computer scientists.”
Congratulations to the authors for a prize well deserved!
At the SIGEVO Meeting at GECCO-2009, the GECCO-99 paper introducing the Bayesian optimization algorithm (BOA) was one of the two papers that received GECCO Impact Award. This is a new award and it focuses on past GECCO papers that have made most impact and have had most citations. The paper was in fact a project […]
At the SIGEVO Meeting at GECCO-2009, the GECCO-99 paper introducing the Bayesian optimization algorithm (BOA) was one of the two papers that received GECCO Impact Award. This is a new award and it focuses on past GECCO papers that have made most impact and have had most citations. The paper was in fact a project from David Goldberg’s class Genetic Algorithms in Search, Optimization, and Machine Learning (GE-485) at the University of Illinois at Urbana-Champaign.
The two awarded papers were:
The second leg of the 2009 Simulated Car Racing Championship just ended. We received around ten submissions. Five new submissions plus the submissions from CEC-2009. Three participants of the CEC-2009 simulated car racing competition updated their drivers.
The three tracks used for the GECCO-2009 leg are: Dirt3, E-road and Alpine.
The results of the first qualifying stage […]
The second leg of the 2009 Simulated Car Racing Championship just ended. We received around ten submissions. Five new submissions plus the submissions from CEC-2009. Three participants of the CEC-2009 simulated car racing competition updated their drivers.
The three tracks used for the GECCO-2009 leg are: Dirt3, E-road and Alpine.
The results of the first qualifying stage are summarized in the following table:
COBOSTAR is still the fastest controller around but Onieva and Pelta are getting closer and closer. MrRacer was unfortunately disqualified since the controller crashed in one of the tracks. At the end of the first stage eight controllers have been selected (the ones showed in green in the table) and three were eliminated (the red ones in the table above).
In the second stage, for each track, we run eight races with different starting grids and scored the controllers based on their arrival position.
The results are summarized in the following table:
Onieva and Pelta performed really well and actually won this leg of the championship. Congratulations! Their new controller performed really much better than the previous one.
This results reopens the championship since the championship scoreboard has the two teams separated by just few points as shown in this table:
Please, remind that the controller by Luigi, the champion of the CIG2008 competition, appears in the scoreboard but it cannot be awarded with any prize since it belongs to one of the organizing institutions).
I wish to thank all the participants. The next leg will be held during CIG-2009 in Milan and it will be held live so that people will be able to watch an actual race while it is happening.
More news will be posted later.
Below you may find the slides I used during GECCO 2009 to present the paper titled “Data-Intensive Computing for Competent Genetic Algorithms: A Pilot Study using Meandre”. An early preprint in form of technical report can be found as an IlliGAL TR No. 2009001 or the full paper at the ACM digital library
Related […]
Related posts:
Below you may find the slides I used during GECCO 2009 to present the paper titled “Data-Intensive Computing for Competent Genetic Algorithms: A Pilot Study using Meandre”. An early preprint in form of technical report can be found as an IlliGAL TR No. 2009001 or the full paper at the ACM digital library
Related posts:
After coming back from GECCO I just uploaded the last of the NIGEL 2006 talks at LCS & GBML Central. This last talk was by Jaume Bacardit and GBML for protein structure prediction.
Related posts:NIGEL 2006 Part V: Bernardó vs. LanziNIGEL 2006 Part IV: Llorà vs. CasillasNIGEL 2006 Part III: Butz vs. Barry
Related posts:
After coming back from GECCO I just uploaded the last of the NIGEL 2006 talks at LCS & GBML Central. This last talk was by Jaume Bacardit and GBML for protein structure prediction.
Related posts:
This is the last of the NIGEL talks NIGEL 2006 talks. Enjoy this last one
Video
[vimeo clip_id=5065758 width=”432″ height=”320″]
Slides
[slideshare id=1384657&doc=nigel-2006-bacardit-090504154202-phpapp02]