Back

overview-simulation.png
Location
Windsor
Ontario, Canada
Date Posted
27 Apr 2017

Visit website


Type
PhD Project
University of Windsor

Two PhD positions in artificial intelligence applied to ecological modelling

overview-simulation.png
Windsor
Ontario, Canada
27 Apr 2017

NOTE: this position listing has expired and may no longer be relevant!

Position Description


Two PhD positions are available in the Learning and Simulation for Theoretical Biology group (http://sites.google.com/site/ecosimgroup/home) at the University of Windsor. The students will be involved in a multidisciplinary project based on EcoSim, an individual-based evolving predator-prey ecosystem simulation.

We have conceived EcoSim, a versatile simulation platform that has been designed to investigate several broad ecological questions, as well as long-term evolutionary patterns and processes such as speciation and macroevolution. EcoSim uses a modified version of the Fuzzy Cognitive Map (FCM) model adapted for behavioral modeling. The FCM is used as the behavioral model of the agents and is coded in their genome and therefore subject to evolution. This approach offers compactness with a very low computational requirement while having the capacity to represent complex high level notions. Therefore, each agent possesses its unique FCM and the system can still manage several hundreds of thousands of agents simultaneously in the world with reasonable computational requirements. In a typical run, more than one billion of agents can be born and several thousands of species can emerge or become extinct. This tool generates a huge amount of data representing all the events, the mental state and action of every agent saved for every time step of every run. This thorough tracking system allows for a deep statistical analysis of the whole system using several dedicated tools that we have conceived to extract, measure and correlate parameters that could be useful to understand the underlying and emerging properties of such a complex system. This level of detail is the highest advantage of this approach compared to real data gathering which is highly limited by the large spatial and temporal scale involved in ecological questions. All the results we have obtained using it demonstrate high potential for our approach to handle complex ecological questions, showing evolutionary and ecological phenomena and patterns conform to real observations and giving the possibility to investigate many hypothesis in a reasonable amount of time.
This simulation will be the framework for the study of numerous specific ecological questions in collaboration with biologists. For example, this approach can be used to study complex ecological and evolutionary processes such as the species abundance distribution, patterns and rates of speciation, the evolution of sexual and asexual populations, the interaction and diffusion of an invasive species into an existing ecosystem, etc.

Due to the highly interdisciplinary framework of the project we need to find very motivated students prepared to learn numerous concepts coming from biology and philosophy. However, the environment for this project will be very supportive has our team is already composed of computer scientists and biologists. We will also work in close collaboration with biologists from the University of Windsor and the Great Lakes Institute for Environmental Research. Having knowledge in biology would be advantageous. Having a strong experience in machine learning and C++ programming is highly recommended. Scholarship from the University of Windsor and Research Assistantship from NSERC will be offered associated with these positions.

[1] Mashayekhi M., MacPherson B., Gras R., Species-area relationship and a tentative interpretation of the function coefficients in an ecosystem simulation, Ecological Complexity, 19, 84-95, 2014.
[2] Khater M., Murariu D., Gras R., Contemporary Evolution and Genetic Change of Prey as a Response to Predator Removal, Ecological Informatics, 22, 13-22, July 2014.
[3] Mashayekhi M., MacPherson B., Gras R., A machine learning approach to investigate the reasons behind species extinction, Ecological Informatics, 20, 58-66, March 2014.
[4] Golestani A., Gras R., A New Species Abundance Distribution Model Based on Model Combination, International Journal of Biostatistics, 9(1), 33-48, July 2013, doi 10.1515/ijb-2012-0033.
[5] Golestani A., Gras R., Cristescu M., Speciation with gene flow in a heterogeneous virtual world: can physical obstacles accelerate speciation?, Proceedings of the Royal Society B: Biological Sciences, 279(1740), doi: 10.1098/rspb.2012.0466, 3055-3064, 2012.
[6] Mashayekhi M., Gras R., Investigating the Effect of Spatial Distribution and Spatiotemporal Information on Speciation using Individual-Based Ecosystem Simulation, Journal of Computing, 2(1), 98 – 103, 2012.
[7] Gras R., Devaurs D., Wozniak A., Aspinall A., An individual-based evolving predator-prey ecosystem simulation using a Fuzzy Cognitive Map model of behavior, Artificial Life, 15(4), 423-463, 2009.


How to Apply

Contact information: Dr. Robin Gras rgras@uwindsor.ca


Apply online


Upload your CV (optional)

Upload your Cover Letter (optional)


This is to prevent spam: