Last updated December 24, 2003
Click here for list of Completed Ph.D. Theses on Genetic Programming
A Partial List of People who are active in genetic programming
If you'd like to be added to this list of student working on theses (PhD, masters, bachelors) involving genetic programming, please send email to koza@genetic-programming.org with the following information:
m_ahluwa@btc.uwe.ac.uk
University of West of England
Doing my PhD in Genetic Programming at UWE Bristol,UK. This is my final year
and most of my work is based on Coevolving functions. The ADF's evolve in their
respective populations and they are picked using different selection
strategies. these ADF's are classification
trees. Currently i'm looking for a good real world application before I start
writing my thesis.
guto@cos.ufrj.br
Chromatic properties of graph using GA and GP.
mikah@nip.upd.edu.ph
Implement a GA/GP to optimize the configuration of a magnetic field around a
plasma, using a 1d plasma simulation. It's actually for my undergrad thesis but
a fully working version but will also very likely be subject of masters.
kabenson@dera.gov.uk
Workingon MPhil/Ph
Control Theory and Applications Center, Coventry University, UK
Working on Master's thesis research (with Lars Nohle) at Chalmers University
of Technology, Institute of Theoretical Physics, at Gothenburg, Sweden with
academic tutor of Peter Nordin and industrial tutor of Eduard Lukschandl,
Ericsson Hewlett-Packard Telecom.
Thesis work consists of two parts:
1) Parallelisation of the JBGP-system, a system for the induction of
Java Bytecode with GP, result of an earlier Master's Thesis.
2) Application of JBGP on routing problems with telecom networks.
eduard.lukschandl@ehpt.com
daveb@insect.sd.monash.edu.au
Monash University
Just completed my Bachelor of Computing Honours thesis in genetic programming.
Specifically, I was looking at the co-evolution of a population of obstacle
avoidance agents against separate population of obstacle courses. The obstacle
avoidance agents were controlled by programs with constructs based on the work
by Craig Reynolds. The `drawing programs' were used to create obstacle courses,
and contained the constructs prog4, penUp, penDown, left, right, up, down. I
was looking at the `robustness' of coevolved obstacle avoidance agents verses
the robustness of (single population, static obstacle course) evolved agents.
http://www.sd.monash.edu.au/~daveb
Completed
E.Chattoe@soc.surrey.ac.uk
Working on PhD
Sub Faculty of Economics, University of Oxford
Evolution of pricing strategies by firms in an oligopolistic market.
http://www.soc.surrey.ac.uk/staff/edmund_chattoe.html
dain@atc.boeing.com
National Technological University
Masters thesis involving GP for mobile robotics. My research is on the
coevolution of control strategies and simulated test environments. Learning
occurs in three iterative phases:
1) Evolve a control strategy to operate in a simulated environment. 2) Test the
control strategy on a real robot in a physical environment and collect
performance data. 3) Evolve a simulated environment based on the control
stragegy and real-world performance feedback.
A.G.Deakin@csc.liv.ac.uk
Working on a PhD (part-time) involving genetic programming (since October
1995). The general area of the research is the evolution of plans and
strategies for problem-solving with GP, with initial focus in pursuit and
evasion scenarios.
http://www.csc.liv.ac.uk/~anthony
maug@copa.rio.com.br
--university???---
Application
of genetic programming in nuclear engineering
problems (fault diagnosis and some symbolic
regression
Masters thesis. February 1997
ekart@lutra.sztaki.hu
Hungarian Academy of Sciences
Doing my PhD thesis in GP at the Computer Science and Automation Research
Institute of the Hungarian Academy of Sciences. I have started my work in
September 1996, so I'm new in the field.
September 1999
noriko-e@is.aist-nara.ac.jp
Working on PhD
Ph.D. research is a robust machine learning and planning agaist a dynamic
environment based on a classifier system and genetic programming. Two
microrobot "Khepera" are tools to verify my research. Although I have
studied a classifier system for my master's thesis, it is my first challenge to
develop an adaptive planning by a genetic programming. If possible, I would
like to exchange a research progress each other.
Noriko Etani (doctoral student)
Artificial Intelligence Laboratory
Graduate School of Information Science
Nara Advanced Institute of Science and Technology
Address: 8916-5, Takayama, Ikoma, Nara 630-01, JAPAN
Phone : +81-7437-2-5265
Fax : +81-7437-2-5269
e-mail : noriko-e@is.aist-nara.ac.jp
http : //ai-www.aist-nara.ac.jp/doc/people/noriko-e/home.html
mark@oresteia.demon.co.uk
De Montfort University
Currently researching for an MPhil / PhD at De Montfort University (Leicester,
England), under the supervision of Dr
Michael Odetayo. I am looking at the use of GA/GP in the prediction of certain
aspects of computer system performance.
ffontana@st.com.uy
Universidad Catolica del Uruguay
Student from Montevideo - Uruguay and now I'm writing my thesis in order to
obtain the degree of "Engineer in Computer Science" or in spanish
"Ingeniero en Informatica." The thesis objetive is to make an
analysis and investigate how can I apply genetic programming techniques to
compress an image with loss of data. The degree is not a master neither PHd, is
only the final requirement to finish my career of 5 years at the
"Universidad Catolica del Uruguay".
anne.gerodolle@cnet.francetelecom.fr
Obtained 1998 a french "DEA" in cognitive science (similar to a MSc).
Defended in June 1998 a thesis about "GP and Robot Control". It is
written in French.
The topic was mainly a review of evolutionary computing, GP and evolutionary
robotics. I applied tree GP, augmented with Rosca's ARL method, to produce
simple behaviours for a Khepera robot (like obstacle avoidance or light
seeking), the objective function being computed by allowing the program to
control the robot for a given time and measuring global criteria like the
distance from the goal. The terminals were the raw values given by all eight
InfraRed Sensors. The document is available at my WWW home page :
http://perso.wanadoo.fr/agerodol
tel : +33 (0) 4 76 76 41 78
jono@cse.ucsc.edu
University of California - Santa Cruz
Working on a Master's Thesis at UCSC. My work involves using evolutionary
methods for computer animation, specifically automatic motion generation in a
physically simulated environment.
M.S.Grant@hw.ac.uk
Working on PhD / M.Sc done and PhD expected in 1999
Heriot-Watt University
M.Sc Comparison of standard GP, ADFs, Ryan's Pygmy Algorithm and Perkis' Stack
GP and these in conjunction with each other. Ph.D. Using GP for Sensor Planning
for Machine Version.
http://www.cee.hw.ac.uk/~msgrant
Michael S. Grant
B.A.Hons (Cantab), M.Sc.
Department of Computing and Electric Engineering
Heriot-Watt University
Edinburgh EH14 4AS
United Kingdom
Tel: (0131) 451 3432 ext. 4174
helen@daimi.aau.dk
Aarhus University
Doing a PhD involving, at the moment, the use of GP in analysing Nuclear
Magnetic Resonance Spectroscopy data, specifically for classification. I will
do some comparative studies with other methods (eg NN), but at present GP is
the main focus.
http://www.daimi.aau.dk/~helen
L.B.Hirsch@shu.ac.uk
Sheffield Hallam University
Now beginning a Ph.D. in which I hope to use GP to model sensorimotor
intelligence. Just finished Masters Thesis which implemented a number of
experiments to model the emergence of cooperative behavior among unrelated
competing organisms using the Iterated Prisoner's Dilemma game and GP.
dale.hooper@sdl.usu.edu
Utah State University
This is a Masters thesis concerning enhancements to GP.
March 1997
Kim.Jansson@vtt.fi
Planning to start a research project, that my lead to a PhD thesis on
optimisation of the manaufacturing of steel parts including scheduling and
nesting problems. The idea is to use Evolutionary Computation for this and at
least for the nesting part use GPs.
http://www.vtt.fi/aut/tau/tau-staf.htm
University of Louisville
dave@archangel.spd.louisville.edu
http://archangel.spd.louisville.edu/~dave
My area of interest is image processing and computer vision, specifically in
the field of fractal analysis. I have written an GA to approximate our fractal
error algorithm and I am working on writing a GP to evolve a better approximation.
mda94hjo@student.hk-r.se
Working towards Bachelor Degree in Computer Science with emphasis on Work
Science
University of Karlskrona / Ronneby
Department of Computer Science And Business Administration
Brief description of topic: To enhance computer opponent AI in games (leisure)
using Genetic
Programming and to investigate means of measuring AI in games.
My WWW home page:
http://www.student.hk-r.se/~mda94hjo/
Thesis page:
http://www.student.hk-r.se/~mda94hjo/thesis/
ykapadia@cs.wright.edu
Wright State University
Working on "Analysis of wrist X rays using Genetic / Evolutionary
Programming Techniques" for my Master's thesis at Wright State University,
Dayton, Ohio, under Dr. Mateen Rizki.
Masters
hooman@cyberdude.com, hkatirai@engmail.uwaterloo.ca
Candidate for Bachelor of Applied Science
University of Waterloo, Ontario, Canada
Department of Computer and Electrical Engineering
Autonomous Classification of Web Pages using Genetic Programming
http://www.eng.uwaterloo.ca/Student/hkatirai
hooman@cyberdude.com, hkatirai@engmail.uwaterloo.ca
Simon.Kent@brunel.ac.uk
Working on PhD
Department of Information Systems and Computin
Brunel University
Uxbridge, Middlesexg
Application of GP and GA to path planning
http://www.brunel.ac.uk/~cspgssk
Simon.Kent@brunel.ac.uk
mario.koeppen@ipk.fhg.de
--university???---
Working on my PhD thesis on the application of softcomputing in image
processing. The research is supported by the german DFG (German Research
Society). Genetic programming is the basic technology for the development of a
system for the automated design of image processing algorithms, which is
presented in the PhD thesis.
http://lucy.ipk.fhg.de/~mario
PhD
ibrahim@cogs.susx.ac.uk
University of Sussex
Will soon be submitting my thesis on "Evolutionary Generalisation and hard
learning problems". A substantial part of the thesis is devoted to GP.
http://www.cogs.susx.ac.uk/users/ibrahim/
tlo@dinf.vub.ac.be
Free University of Brussels
Currently doing research for my PhD thesis. My research can be summarized by
the following keywords : data mining , classifier systems, genetic programming,
machine learning.
http://dinf.vub.ac.be/asg
jli@essex.ac.uk
University of Essex
Using GP on financial prediction.
Biomedical Engineering Group
CEFET-PR / CPGEI
http://www.cpgei.cefetpr.br/~marchesi/Bruno Marchesi
http://www.cpgei.cefetpr.br/~marchesi
marchesi@cpgei.cefetpr.br
Thesis using GP to evolve an expression capable of recognizing based on
features, epileptiform activity in electroencephalography signals. Keywords:
genetic programming, electroencephalography, pattern recognition.
tlm113@mail.USask.Ca
Working on thesis.
ben@chem.eng.usyd.edu.au
University of Sydney
Use of both neural networks and genetic programming for the control relevent
modelling of process systems (ie. in the chemical and allied industries).
www: ftp://129.78.12.5/www/pgrad/ben2.htm
Just finishing up my PhD
degree: PhD
EEBIC group, Department of Computer Science, Univeristy of Birmingham, U.K.
Topic: "Natural Selection in Evolutionary Computation".
Working under Aaron Sloman and Riccardo Poli on the notion of 'fertility' of a data structure or program subtree. Individuals in a GP may have the same fitness, but their lifetime reproductive success differs. Some of the differences are random, but I am investigating the possibility that some of the differences are non-random. For want of a better term, I call this 'fertility'.
email: idm@cs.bham.ac.uk
aler@inf.uc3m.es
Working towards a PhD degree.
Universidad Carlos III de Madrid, Spain
Computer Science Department.
Working in a multistrategy learning approach to obtain
control-knowledge for problem-solving (planning). Two methods (Hamlet and
EvoCK (GP based)) for obtaining control rules are combined, the purpose
being to easily add useful biases to Hamlet by means of the GP paradigm. In GP,
biases can be declaratively stated very easily in a fitness function. Viewing
it from another point of view, I am interested in improving GP with background
knowledge obtained by other machine learning
techniques. The evolving structure (the program) is a set of control rules
whose left
hand side allows for relational representations and matching with
variables, and whose right hand side influences the state of a planning
system (PRODIGY). So, although it's not a generic typical production
system, it's rather close.
http://grial.uc3m.es/~aler
naka@tyl.t.u-tokyo.ac.jp
KHC02440@niftyserve.or.jp
University of Tokyo
Application of GP as my Phd thesis. My paper is in the
Late-Breaking Papers at GP 96 pages133-140 and its title is "Capturing
Preference into a Function Using Interactions with a Manual Evolutionary Design
Aid System"
http://woodstock.t.u-tokyo.ac.jp/~naka/Nakanishi.html
March 1998
Working on Master's thesis research (with Henrik Borgvall) at Chalmers
University of Technology, Institute of Theoretical Physics, at Gothenburg,
Sweden with academic tutor of Peter Nordin and industrial tutor of Eduard
Lukschandl, Ericsson Hewlett-Packard Telecom.
Thesis work consists of two parts:
1) Parallelisation of the JBGP-system, a system for the induction of
Java Bytecode with GP, result of an earlier Master's Thesis.
2) Application of JBGP on routing problems with telecom networks.
eduard.lukschandl@ehpt.com
chocron@robot.uvsq.fr
Laboratoire de Robotique de Paris
Evolving dynamically reconfigurable modular robots to meet evolutive task
specifications. Adapt GP for evolving behavior of a walking robot as well as
its kinematic topology.
PhD work (2nd year)
michael.oneill@ul.ie
Department of Computer Science and Information Systems
University of Limerick
Ireland
Currently doing research for a PhD in the area of Grammar based
Program Induction with Genetic Algorithms and Genetic Programming.
michael.oneill@ul.ie
http://shine.csis.ul.ie/
norman@dcs.st-and.ac.uk
University of St Andrews, Scotland
GP with a genotype - phenotype distinction, so that (for example) the language
of the resulting programs is not hard-wired into the GP engine.
http://www.dcs.st-and.ac.uk/~norman/
Doing a PhD in GP, but have not
started working on the actual thesis yet
S.Perkins@ed.ac.uk
Working on PhD
Simon Perkins
Department of Artificial Intelligence
Edinburgh University
Using a variant of GP in conjunction with shaping
techniques to incrementally evolve controllers for active vision
behaviours such as light tracking, motion tracking and target seeking.
Tel: +44 131 650 4496
http://www.dai.ed.ac.uk/daidb/homes/simonpe/
j.curzon-price@ic.ac.uk
Imperial College
Using GP/GA for a Phd thesis applying the techniques to industrial organisation
theory & practice, using co-evolutionary set-ups to simulate strategic
behaviour in not-very-competitve markets (UK electricity pool, European natural
gas markets).
http://is.eunet.ch/Customers/curzon
1998
wrp103@psu.edu
wrp@ks.unisys.com
Penn State University
Method to evolve structured programs using GP. More information on my home
page,
http://www.gv.psu.edu/personal/wrp103/home.htm
Expect to finish in 1997
J.Pujol@cs.bham.ac.uk
University of Birmingham
Optimisation of the topology and
weights of NNs using a GP paradigm derived from Poli's PDGP (Parallel
Distributed GP, see technical report at URL:
ftp://ftp.cs.bham.ac.uk/pub/tech-reports/1996/CSRP-96-15.ps.gz. PhD student
working on GP for the optimisation of neural networks under Prof. R. Poli.
A.M.Radi@cs.bham.ac.uk
University of Birmingham
Working on discovering new learning rules for feed-forward neural nets using
standard GP. PhD student working on GP for the optimisation of neural networks
under Prof. R. Poli.
randwill@cps.msu.edu
Michigan State University
Am an undergraduate doing research in the area of genetic programming.
Attempting to model pack/herd behavior on a simple genetic programming system
known as lil-gp. I have already developed a basic hunter/evader that works well
for two individuals but am attempting to now use multiple hunters and evaders
using ADFs.
517-355-7035
gemmac@salleURL.edu
Universitat Ramon Llull
Working on PhD
Working on genetic programming applied to acoustic modelation of street noise
Gemma Calzada i Roca
Departament d'Inform`tica
Enginyeria La Salle
Universitat Ramon Llull
Passeig Bonanova,8
08022 Barcelona
Spain
Tele: +34 932 902 423
http://www.salleURL.edu/~gemmac
MS in Computer Engineering
Department of Computer Engineering
King Fahad University of Petroleum and Minerals
Dhahran, Saudi Arabia
Working on Test Generation for Sequential circuits using evolutionary / genetic techniques
g980329@kfupm.edu.sa
shsham@essex.ac.uk
University of Essex
Working towards my PhD and my research involves the application of genetic
programming to certain aspects of communication networks.
PhD student
jsherrah@eleceng.adelaide.edu.au
University of Adelaide
I am starting my third year of a (hopefully) 3 year Ph.D. program at the
University of Adelaide, South Australia. The topic is genetic programming for
pattern recognition. In short, pattern classification involves many arbitrary
and ad-hoc decisions, such as feature selection and extraction, and the choice
of classifier. I am using genetic programming to automatically determine which
features and classifier are good for a given problem.
http://www.eleceng.adelaide.edu.au/Personal/jsherrah/index.html
March 1998
hussein@yoda.udw.ac.za
University of Durban Westville
Working on MSc thesis
Almost finished with my MSc thesis entitled "Genetic Programming in
Mathematica" which investigates the feasability of serial and parallel
implementations which take advantage of Mathematica's modelling abilities.
biso@cheerful.com
biso@usa.net
Post graduate student at the Faculty of Engineering Cairo University preparing
a thesis about GP and GA.
talko@cs.mu.oz.au
University of Melbourne
Using GP to evolve modular (and possibly recurrent) neural networks with
different neurons having (possibly) different activation functions.
MSc student
jens@rt.e-technik.tu-darmstadt.de
Working on Double Masters Degree in Electrical Engineering and Managment
(Wirtschaftsingeni
eurwesen)
Darmstadt University of Technology, Germany
Control Systems Theory and Robotics Group
(http://www.rt.e-technik.tu-darmstadt.de)
Automatic Controller Design by Means of Genetic Programming. As followup to
some previous work done at this group, I'll focus on the question how
to automatically assess the quality of a controller and on the application
to multi-variable nonlinear dynamic processes.
(see http://www.rt.e-technik.tu-darmstadt.de/~jens/project.html)
Faculty advisor: Peter
Marenbach
trenaman@cs.may.ie
Maynooth College
This thesis investigates the use of concurrency in genetic
programming to evolve agents controlled by populations of individuals whose
emergent interactions create the behaviour of the agent. Using concurrent
genetic programming, such agents have been evoled using different memory
architectures (shared and unshared memory) and under different scheduling
strategies (fine.coarse-grained paralellism, evolved concurrency). Results
to date indicate that such agents exploit the concurrent execution to
effect internal state corresponding to models of the environment, and yield
individuals of very low complexity. Current work aims to apply this
technique to evolve controllers for khepera robots in environments where
internal models of reality are essential for high fitness. Papers relating
to this work appear in GP97 (late-breaking papers), AI97 (conf. proc.),
GP98 (conf proc. and late-breaking papers) and can be downloaded from the
web-site below. The current completion date is set for December 1998.
Adrian Trenaman
Department of Computer Science
NUI Maynooth
Co. Kildare
Ireland
http://www.cs.may.ie/~trenaman/
fferveg@alcazaba.unex.es
Facultad de Biblioteconomma y Documentacisn
Universidad de Extremadura
GP and GA in PhD thesis.
A.Watson@plymouth.ac.uk
University of Plymouth
Third (final) year PhD student researching "Calibrating Gas Turbine Design
Software Using Genetic Programming." Part funded by Rolls-Royce Plc.
http://www.tech.plym.ac.uk/soc/research/edc
October 1997
Working on PhD
School: School of Civil and Structural Engineering
University: Nanyang Technological University, Singapore
Topic: GP-based approach for structural optimization and design
P144191201@ntu.edu.sg
g3258501@grad.cc.nccu.edu.tw
National Chengchi University
PhD thesis about the application of genetic programming in economics. These
topics are econometrics, game theory, and economic environment based on
multiagent framework.
http://econo.nccu.edu.tw/
T.Yu@cs.ucl.ac.uk
University College London
Ph.D. student doing research in applying Functional Programming techniques to
enhance GP performance. These include: 1. evloving expression-based programs
(lambda caculus); 2. strongly typed GP using a formal type system which
supports parametric polymorphism, type inheritance and user-defined type; 3.
crossover on partial application function nodes; 4. higher-order functions
salem@ele.puc-rio.br
Catholic University of Rio
Evolvable Hardware and it will possibly involve Genetic Programming.
· The home page of Genetic Programming Inc. at www.genetic-programming.com.
· For information about the field of genetic programming in general, visit www.genetic-programming.org
· The home page of John R. Koza at Genetic Programming Inc. (including online versions of most papers) and the home page of John R. Koza at Stanford University
· For information about John Koza’s course on genetic algorithms and genetic programming at Stanford University
· Information about the 1992 book Genetic Programming: On the Programming of Computers by Means of Natural Selection, the 1994 book Genetic Programming II: Automatic Discovery of Reusable Programs, the 1999 book Genetic Programming III: Darwinian Invention and Problem Solving, and the 2003 book Genetic Programming IV: Routine Human-Competitive Machine Intelligence. Click here to read chapter 1 of Genetic Programming IV book in PDF format.
· For information on 3,198
papers (many on-line) on genetic programming (as of June 27, 2003) by over 900
authors, see William
Langdon’s bibliography on genetic programming.
· For information on the Genetic Programming and Evolvable Machines journal published by Kluwer Academic Publishers
· For information on the Genetic Programming book series from Kluwer Academic Publishers, see the Call For Book Proposals
· For information about the annual Genetic and Evolutionary Computation (GECCO) conference (which includes the annual GP conference) to be held on June 26–30, 2004 (Saturday – Wednesday) in Seattle and its sponsoring organization, the International Society for Genetic and Evolutionary Computation (ISGEC). For information about the annual Euro-Genetic-Programming Conference to be held on April 5-7, 2004 (Monday – Wednesday) at the University of Coimbra in Coimbra Portugal. For information about the 2003 and 2004 Genetic Programming Theory and Practice (GPTP) workshops held at the University of Michigan in Ann Arbor. For information about Asia-Pacific Workshop on Genetic Programming (ASPGP03) to be held in Canberra, Australia on December 8, 2003. For information about the annual NASA/DoD Conference on Evolvable Hardware Conference (EH) to be held on June 24-26 (Thursday-Saturday), 2004 in Seattle.