9 July 2012

GECCO-2012: Three EDA sessions, and more

Getting ready to start day 3 at GECCO-2012. Today, there are three sessions with EDA papers in them, so I hope to come to all of these. I’m also giving my second presentation in one of the sessions on distance-based bias in model-directed optimization of additively decomposable problems. The basic idea is to mine models from prior runs of model-directed optimizers for information about the problem domain to introduce bias when solving new problem instances of similar type, so that these instances can be solved with increased speed, accuracy, and reliability. Essentially, this allows an optimizer to learn from solving a certain type of problems to get better at solving similar problems in future. In this specific approach, models from prior runs are combined with a distance metric based on problem-specific knowledge. For more information about this work, you can download MEDAL Reports 2012001 and 2012004.

8 July 2012

GECCO 2012: Getting ready for the opening reception and the EDA day

After a full day, starting with my EDA tutorial, followed by a number of interesting discussions, I am now getting ready for the opening reception. Tomorrow is the EDA day (EDAy) for me with all the EDA-related sessions. Will be busy but fun.

8 July 2012

GECCO-2012 continues, now getting ready for the tutorial on EDAs

Day 2 of GECCO-2012 is just starting, getting ready to give the tutorial Probabilistic Model-Building Genetic Algorithms at 10:40am.

7 July 2012

GECCO-2012: Attending the tutorial on evolution strategies

I’ve just registered at GECCO-2012 in Philadelphia, and am attending my first tutorial Evolution Strategies by Thomas Baeck.

24 May 2012

STARS in MEDAL


MEDAL is going to participate in the STARS program this year. STARS (Students and Teachers As Research Scientists) provides an opportunity for best high school students and also their teachers to participate in research projects with university faculty and graduate students. If you want to learn more about the program, you can look up this article in UMSL Daily. In any case, this is the first time we participate in the program and we’re quite excited about it; it should be fun.

17 May 2012

Google’s PageRank algorithm finds cancer biomarkers

Medical News Today reported on an interesting study done by researchers from the Dresden University of Technology, who proposed the use of the Google’s PageRank algorithm for identification of new cancer biomarkers, which are molecules produced by cancer cells.

20 April 2012

Inaugural class of NAI members from UMSL inducted earlier today

UMSL NAI Chapter Inauguration Photo
I’ve just come back from the event that launched the University of Missouri-St. Louis chapter of the National Academy of Inventors (NAI). During the event, I was inducted as a member of the inagural class into this NAI chapter, thanks to the patent Method for Optimizing a Solution Set that we filed with David E. Goldberg during my Ph.D. study at the University of Illinois at Urbana-Champaign.

14 April 2012

University of Missouri researchers develop a new biological data analysis tool


Missourian has just reported that a computer algorithm developed by University of Missouri (MU) researchers “has the potential ability to create more effective medicines and detect side effects of drugs.” From the article:


The technology can scan all human proteins for sequence regions that are identical among people, said Dmitry Korkin, an MU professor in computer science. Finding these regions is useful in determining whether a certain drug that targets a region in one protein can target the same region in other proteins, which are otherwise not related, Korkin said. “If that happens, we can have either a side effect or potentially have another therapeutic effect,” Korkin said. Korkin led interdisciplinary research using this new tool to find identical DNA sequences in plants.

The paper describing the research was published in the Proceedings of the National Academy of Sciences.

12 April 2012

New MEDAL reports available

We are pleased to announce the following MEDAL technical reports published so far in 2012:

MEDAL Report No. 2012006
Influence of selection on structure learning in Markov network EDAs: An empirical study
Alexander E.I. Brownlee, John A. W. McCall, and Martin Pelikan  (2012)
Abstract
Download PDF

MEDAL Report No. 2012005
Linkage learning using the maximum spanning tree of the dependency graph
B. Hoda Helmi, Martin Pelikan, and Adel Rahmani  (2012)
Abstract
Download PDF

MEDAL Report No. 2012004
Transfer learning, soft distance-based bias, and the hierarchical BOA
Martin Pelikan and Mark W. Hauschild  (2012)
Abstract
Download PDF
Download PS

MEDAL Report No. 2012003
Introduction to estimation of distribution algorithms
Martin Pelikan, Mark W. Hauschild, and Fernando G. Lobo  (2012)
Abstract
Download PDF
Download PS

MEDAL Report No. 2012002
Image Segmentation using a Genetic Algorithm and Hierarchical Local Search
Mark W. Hauschild, Sanjiv Bhatia and Martin Pelikan  (2012)
Abstract
Download PDF
Download PS

MEDAL Report No. 2012001
Distance-based bias in model-directed optimization of additively decomposable problems
Martin Pelikan and Mark W. Hauschild  (2012)
Abstract
Download PDF
Download PS

11 April 2012

Cooperation leads to intelligence

Discovery News reported today on a study that argues that cooperation is a driving force for the evolution of intelligence. The study was done by Luke McNally and Andrew Jackson from the Trinity College Dublin, and Sam Brown from the University of Edinburgh. A genetic algorithm was used in the study to simulation evolution. From the original article:


Cooperation leads to intelligence. In fact, researchers believe that Intelligence, in this case, correlated with the size of a brain’s neural network, in our species and others may have been an adaption for tool use, for social learning, and for the accumulation of culture. Intelligence might also be the result of sexual selection.