NCSI

   

Large Scale Genetic Programming for Scientific Image Understanding (Learner Position)


Shodor > NCSI > XSEDE EMPOWER > XSEDE EMPOWER Positions > Large Scale Genetic Programming for Scientific Image Understanding (Learner Position)

Status
Completed
Mentor NameDirk Colbry
Mentor's XSEDE AffiliationCampus Campion
Mentor Has Been in XSEDE Community4-5 years
Project TitleLarge Scale Genetic Programming for Scientific Image Understanding (Learner Position)
SummaryWe are looking for students interested joining our research team to help develop software to scale a prototype Simple Evolutionary Exploration (SEE) library to utilize large scale computing systems. The search space involved in this research is extremely large and requires massive computing resources. This research will look into leveraging High Performance Computing Resources (XSEDE), HTCondor and Cloud Resources. The long term goal of the project is to build image annotation system that works in "real time" with the researchers to explore the algorithm space for solutions to scientific image understanding problems.
Job DescriptionStudents will work with a research team continue to build and scale a prototype software library to run on very large systems. The current prototypes are written in Python and searches the "algorithm space" of image segmentation algorithms using single sample learning. The nature of the search space is very large and highly non-differentiable so standard optimization techniques do not apply. However, this software utilizes Genetic Programming which is pleasantly parallel and therefore this algorithm can easily leverage large scale systems.
Computational ResourcesThe current prototype has been run on a local Tear-III XSEDE resource and multiple cloud resources (Azure, Google Cloud and AWS). We hope to further test the system on these resources and specifically build an interface that works with HTCondor and maybe XSEDE cloud resources such as Jetstream.
Contribution to CommunityThe tools we are developing should have a broad impact for scientific image understanding. our hope is to build web-based gateway which will allow researchers to annotate their images on the front end while the tool searchers for an automated solution on the back end.
Position TypeLearner
Training PlanLearning goals will be established individually for each student who will work in a structured research group along side other undergraduates. New students will learn how to use GIT, Python, Jupyter notebooks, unit testing, linting and other software development tools. Individual projects are customized based on students interests and experience.
Student Prerequisites/Conditions/QualificationsAlthough no prior work experience is required, some knowledge of computer programming (primarily Python) or scientific computing is expected. Ideally, applicants will also have experience in one or more of the following: scientific image understanding, using HPC systems, hacking or tinkering. The research is gaining a lot of momentum and there are lots of opportunities.
DurationSemester
Start Date08/01/2021
End Date12/15/2021

Not Logged In. Login