Tomasz Oliwa

About me   
Publications    Presentations   Projects    Teaching    Affiliations    Other Interests



tomasz oliwa
Postdoctoral Scholar

Toyota Technological Institute at Chicago
6045 S Kenwood Ave
Chicago, IL 60637

E-mail: oliwa at {ttic dot edu | uchicago dot edu | uga dot edu}
           zophar at uni-koblenz dot de
GPG public key available: n.pubkey.asc.txt

Phone: +1-773-834-2622
ttic



About me

Tomasz Oliwa

I am a
postdoctoral scholar at the Toyota Technological Institute at Chicago (TTIC), located on the
campus of the University of Chicago. My work under Dr. Yang Shen involves modeling,
dimensionality reduction, optimization and machine learning using HPC in the context of protein docking.

I received my Ph.D. in Computer Science from the University of Georgia, Athens, USA.
My work and research was with the Evolutionary Computation & Machine Learning (ECML) Lab.

In addition to that, I was also graduate student at the Institute for Artificial Intelligence under a scholarship
of the Deutscher Akademischer Austausch Dienst (DAAD).

I studied Computer Science with a focus in Artificial Intelligence and a minor subject in Physics
at the University of Koblenz-Landau, Germany and finished with the degree Diplom-Informatiker.


International Conference Publications (Peer reviewed)

2014

Tomasz Oliwa and Yang Shen, "A framework of complex-based normal mode analysis (cNMA) to model
conformational changes of proteins upon interactions", 2014 IEEE International Conference on Bioinformatics
and Biomedicine (
Submitted, IEEE BIBM 2014)

Tomasz Oliwa and Yang Shen, "A complex-based normal mode analysis for conformational changes upon
protein binding", 3DSIG 2014 Poster Abstract, Structural Bioinformatics and Computational Biophysics,
ISMB Satellite Meeting, Boston, USA, 2014.


2013

Tomasz Oliwa and Khaled Rasheed, "An overlapping variable linkage benchmark suite".
In Proceeding of the fifteenth annual conference companion on Genetic and evolutionary computation conference
companion (GECCO '13 Companion), Christian Blum (Ed.). ACM, New York, NY, USA, 127-128.
BibTex - ACM Link - ACM Author-Izer PDF

2012

Tomasz Oliwa and Khaled Rasheed, "A Surrogate-Assisted and Informed Linkage Aware GA",
Proceedings of the fourteenth international conference on Genetic and evolutionary computation conference
companion (GECCO Companion '12), Terence Soule (Ed.). ACM, New York, NY, USA, 1467-1468.
BibTex - ACM Link - ACM Author-Izer PDF

2011

Tomasz Oliwa and Khaled Rasheed, "A surrogate-assisted linkage inference approach in genetic algorithms",
Proceedings of the 13th annual conference on Genetic and evolutionary computation (GECCO 2011),
Natalio Krasnogor (Ed.). ACM, New York, NY, USA, 997-1004.
BibTex - ACM Link - ACM Author-Izer PDF

2010

Tomasz Oliwa and Khaled Rasheed, "A Machine Learning Approach for Sensitivity Inference in Genetic Algorithms",
Proceedings of the 2010 International Conference on Genetic and Evolutionary Methods (GEM'2010),
Las Vegas, USA, 2010.
BibTeX

2008

Tomasz Oliwa, "Genetic Algorithms and the abc Music Notation Language for Rock Music Composition",
Proceedings of ACM Genetic and Evolutionary Computation Conference 2008 (GECCO 2008),
ACM, New York, NY, 1603-1610.
BibTeX - ACM Link - ACM Author-Izer PDF - Composed music as MIDI and MP3

Tomasz Oliwa and Markus Wagner, "Composing Music with Neural Networks and Probabilistic Finite-State Machines",
Proceedings of the Sixth European Workshop on Evolutionary and Biologically Inspired Music, Sound, Art and Design
(EvoMUSART 2008), Springer Berlin / Heidelberg, Springer, 503-508.
BibTeX - Springer LNCS 4974 - PDF - Composed music as MIDI and MP3
The final publication is available at www.springerlink.com.


Other Conference/Symposium Presentations

2014

Presentation
Tomasz Oliwa and Yang Shen, "On conformational changes associated with protein-protein interactions:
Normal mode analysis for encounter complexes", presented at the Protein Folding Conference 2014,
Punta Cana, Dominican Republic,
July 2014

2013

Presentation
Tomasz Oliwa, "Introduction to Evolutionary Computation", presented at the Center for Simulational Physics
seminar, Athens, GA, USA, March 2013

2012

Presentation
Tomasz Oliwa and Khaled Rasheed, "A surrogate-assisted and informed linkage aware GA",
presented at the Computer Science Advisory Board Poster Session, Athens, GA, USA, October 2012

2010

3rd place: Simulation Soccer Championship
"Bronze" Place Winner Award for developing a multi-agent virtual soccer team and competing in the Annual 2010
Simulation Soccer Championship in CSCI 8220 Advanced Simulation Systems.

2009

1st place: Best Poster Award (from 41 accepted posters)
Best Poster Award, Tomasz Oliwa, "Music Composition with Artificial Intelligence", poster presentation at the
UGA Computer Science Research Day, Athens, GA, USA.

Presentation
Tomasz Oliwa, "Machine Learning techniques utilized for Prediction and Classification on the Alzheimer's
Disease Neuroimaging Initiative (ADNI) dataset", presented at the Graduate Student Conference, Athens, GA, USA.

2008

Presentation
Tomasz Oliwa, "Introducing Structures for Music Composition with Probabilistic Finite-State Machines", presented at the
8th Annual Georgia Graduate Student Interdisciplinary Conference (GGSIC 2008), Athens, GA, USA.

2006

Presentation
Tomasz Oliwa, "Understand the Java Memory Model", presented at the 5th KeY Symposium 2006, Speyer, Germany.


Selected Projects

Protein Docking
My interests are normal mode analysis, dimensionality reduction, optimization, machine learning and HPC in the context of protein docking.

Machine Learning
I am interested in developing and applying machine learning techniques in various domains:

• Problem structure learning with surrogate models (e.g. SVMs, neural networks) in the context of global optimization and
assistance of evolutionary algorithms for significant performance and result quality improvements.

• Music composition based on inductive learning with time-delay neural networks, you can get an overview by looking at this poster.

• Multi-agent predator-prey simulators with feed-forward neural network trained agents.

• Multi-model based predictors of Alzheimer's disease based on medical data sets (NIH Challenge Grant proposal), aim presented here.

Evolutionary Computation
• Currently I am investigating algorithms to detect and characterize linkage and sensitivity in the domain of evolutionary algorithms.

• Evolutionary creativity in the domain of music composition. You can take a look at my publication for a system based on genetic algorithms
which can compose multi-instrumental music or listen to some of composed music by yourself here.

• Implementation of HPC (MPI, OpenMP) and GPGPU optimization techniques (CUDA, OpenCL).

Formal Methods, Verification
At the University of Koblenz-Landau, Germany I was a graduate research assistant in the KeY Project, a formal methods
tool for specification and verification of object oriented software.

Virtual/Simulated GNU/Linux Networks
During my project internship at the University of Koblenz, Germany I worked on an implementation of NIS and NFS
in Virtual Network User Mode Linux (VNUML).


Teaching

I have held guest lectures in the following courses:

Fall 2013

CSCI/ARTI 8950 Machine Learning

Spring 2013

CSCI/ARTI 8950 Machine Learning

Fall 2012

CSCI 4560/6560 Evolutionary Computation

Fall 2011

CSCI 4560/6560 Evolutionary Computation


Professional Affiliations

• Program committee member of the Genetic and Evolutionary Computation Conference 2014 (GECCO 2014)
• Program committee member of the Genetic and Evolutionary Computation Conference 2013 (GECCO 2013)
• Program committee member of the Genetic and Evolutionary Computation Conference 2012 (GECCO 2012)
• Reviewer for the IEEE Congress on Evolutionary Computation 2014 (CEC 2014)
• Reviewer for the IEEE Congress on Evolutionary Computation 2013 (CEC 2013)
• Reviewer for the IEEE Congress on Evolutionary Computation 2012 (CEC 2012)
• Reviewer for the IEEE Symposium Series on Computational Intelligence 2013 (SSCI 2013)
• Member of the Association for Computing Machinery (ACM)
• Member of the ACM Special Interest Group on Genetic and Evolutionary Computation (SIGEVO)
• Former member of the Evolutionary Computation & Machine Learning (ECML) Lab in the Computer Science department
of The University of Georgia, Athens, GA
• Former member of the working group Artificial Intelligence (AGKI) in the Computer Science department of the
University of Koblenz-Landau, Koblenz, Germany


Interests outside of Computer Science and Artificial Intelligence

Electric Guitar
Download one of my songs/recordings here. This hard rock instrumental was recorded under GNU/Linux with audacity
for the guitar recording and hydrogen for the drums. To listen to it, you will need the Ogg Vorbis sound lib (Open source).

Karate
I used to workout with the UGA Karate Club and hope to find a good martial arts club at the University of Chicago.

Literature
I enjoy novels written by Stanislaw Lem, Franz Kafka or Thomas Pynchon and I also like experimental novels
like House of Leaves by Mark Z. Danielewski.