Dear colleagues,
It would be greatly appreciated if you could bring the following to the
attention of qualified candidates.
The Dutch national research institute for mathematics and computer science (CWI,
Centrum Wiskunde & Informatica) located in Amsterdam, and the Leiden University
Medical Center (LUMC) together have 10 vacancies for fully funded Ph.D. students
on various joint AI projects on innovating model-based evolutionary algorithms
from algorithmic foundations to medical applications.
https://www.cwi.nl/jobs/vacancies/phd-students-on-innovating-model-based-evolutionary-algorithms-from-algorithmic-foundations-to-medical-applications
Description
-----------
Evolutionary Algorithms (EAs) are metaheuristic search methods. Their underlying
concept of improving multiple solutions at once is a powerful one, making them
popular for tackling complex optimization and machine-learning problems. EAs are
increasingly considered a key ingredient needed to fundamentally advance AI.
The ability to exploit important structural features of a problem can make the
difference between obtaining efficient, i.e., low-polynomial, and inefficient,
i.e., exponential, scale-up of the time required to find the optimum as problem
size increases. At CWI, a dedicated research line exists on the development of
state-of-the-art model-based EAs that automatically detect, learn, and
incorporate problem structure to achieve superior performance. At the forefront
of this research line is the Gene-pool Optimal Mixing Evolutionary Algorithm
(GOMEA) that has been shown to be capable of solving single- and multi-objective
optimization benchmark problems with millions of variables on ordinary desktop
computers.
GOMEA further has unique properties that lend itself to achieve excellent
results for real-world problems. CWI collaborates closely with medical centres
to improve clinical practice, particularly in radiation oncology. Already, this
has led to winning awards from both the EA field as well as the radiation
oncology field and an uptake of software in clinical practice to improve the
design of brachytherapy (internal radiation therapy) treatment plans for the
treatment of prostate cancer.
Building upon these successes, CWI, in collaboration with LUMC, have recently
acquired funding for various projects that together comprise 10 PhD positions.
Briefly, these projects are as follows:
Fast, accurate, and insightful brachytherapy treatment planning for cervical
cancer through AI: 1 PhD student at CWI and 1 PhD student at LUMC (both
positions on extending our award-winning automated brachytherapy treatment
planning approach to cervical cancer, including making novel combinations
with machine learning)
DAEDALUS - Decentralized and Automated Evolutionary Deep Architecture
Learning with Unprecedented Scalability: 2 PhD students at CWI (on designing
GOMEA variants for scalable distributed deep neural architecture search), 1
PhD student at LUMC (on real-world medical applications in medical image
analysis and radiation therapy dose analysis with multiple treatment
centres)
EXAMINE - Evolutionary eXplainable Artificial Medical INtelligence Engine: 1
PhD student at CWI (on designing GOMEA variants for full (deep) neural
network (and autoencoders) evolution and its integration in an explainable
AI engine) [1 PhD student position at LUMC already filled]
TRUST-AI - Transparent, Reliable and Unbiased Smart Tool for Artificial
Intelligence: 2 PhD students at CWI (on innovating the genetic programming
variant of GOMEA for symbolic regression with niching and mixed-constant
evolution and on finetuning the algorithms to work in clinical practice
regarding rare tumors in the head and neck region) [in close collaboration
with LUMC]
MODIR - Multi-Objective Deformable Image Registration: 2 PhD students at
LUMC (on using GOMEA on a GPU to optimize high-dimensional models of tissue
deformations and finetuning algorithms to work in clinical practice for the
pelvic region) [1 postdoc and 1 PhD student position at CWI already filled]
All PhD students are intended to work on and/or with novel GOMEA variants.
However, some positions are more fundamental than others and sometimes there is
room to shift content somewhat between topics. It is vital to understand that in
all cases, a thorough understanding of the underlying algorithmics and their
software implementation is essential. Hence, candidates should ensure that they
have the proper background and motivation when applying. Moreover, as all
positions are part of larger collaborative projects, candidates are expected to
spend at least one day per week at the other institute. Finally, several of the
above projects also have international research partners (in Europe, the US,
Australia, and New Zealand). Candidates are therefore expected to be open to
spend time abroad with these research partners.
For more information, including Requirements, Terms and Conditions, the
Application procedure, information about the host institutes and the research
group, please visit the following link:
https://www.cwi.nl/jobs/vacancies/phd-students-on-innovating-model-based-evolutionary-algorithms-from-algorithmic-foundations-to-medical-applications
For more information about the vacancy, please contact Peter A.N. Bosman,
[log in to unmask]
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Prof.dr. Peter A.N. Bosman
Senior Researcher at Centrum Wiskunde & Informatica (CWI)
(Dutch National Research Institute for Mathematics and Computer Science)
Life Sciences and Health Research Group
Professor of Evolutionary Algorithms at Delft University of Technology
Algorithmics Section of the Department of Software Technology
Faculty of Electrical Engineering, Mathematics and Computer Science
E [log in to unmask] W http://homepages.cwi.nl/~bosman T +31(0)205924265
-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
########################################################################
To unsubscribe from the EVOLUTIONARY-COMPUTING list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=EVOLUTIONARY-COMPUTING&A=1
This message was issued to members of www.jiscmail.ac.uk/EVOLUTIONARY-COMPUTING, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|