Profile for David Mathias

David Mathias profile photo

FirstGeneration badge,

Specialty area(s)

Genetic algorithms, evolutionary computation, swarm systems

Genetic algorithms solve problems using the principles of Darwinian evolution.  The algorithm maintains a population of candidate solutions to the problem which it evolves over a number of generations, using algorithmic analogs to reproduction, mutation, and survival of the fittest.  The challenge is to encode the problem in such a way that allows the evolutionary progress to be meaningful and to implement genetic operators that are effective.

A newer area of inquiry for me is decentralized swarm-based systems.  Modeled by natural swarms, such as bees and ants, swarm-based systems use a large number of small, inexpensive computational agents to solve problems.  These agent can be robots, simulations, or other software programs.  The large number of agents in a. swarm provides redundancy and, therefore, protection against agent loss.  The key to making these systems work is an effective mechanism allowing the agents to make independent decisions about how to contribute to the overall goal of the swarm.

Brief biography

Originally from Wilmington, Delaware, I have also lived in St. Louis, MO; Columbus, OH; Blonay, Switzerland; and Tampa, FL, before coming to UWL.  I enjoy mountain hiking, tinkering with cars, and English soccer (er, football).

 

Current courses at UWL

CS 220 Program Design II

CS 224 Programming in Python 

 

Education

D.Sc. in Computer Science: Washington University in St. Louis (1996)

M.S. in Computer Science: Washington University in St. Louis (1993)

B.S. in Computer Science: University of Delaware (1991)

 

Teaching history

Over the past 20 years, I have taught introductory programming courses, operating systems, analysis of algorithms (my favorite), discrete math (a close second), programming languages (my least favorite), and game design.

Professional history

Previous stops in my academic career include the Ohio State University (1997 - 2009), and Florida Southern College (2014 - 2018).  At Florida Southern, I founded the Department of Computer Science -- CS had previously been a major within Mathematics.  In addition, I held the Charles and Mildred Jenkins Endowed Chair in Mathematics and Computer Science and was a member of the Faculty Senate.

 

Research and publishing

The most rewarding aspect of my work is research collaboration with undergraduate students.  Several of my previous undergrads have gone on to doctoral study in CS.  I am currently working with several UWL CS majors, including an Eagle Apprentice.  My current research in swarm systems is funded by the National Science Foundation.

 

Kudos

presented

David Mathias, Computer Science; and Kaelan Engholdt, CS: Computer Science BS; Annie Wu, University of Central Florida; presented "Variable Response Duration Promotes Self-organization in Decentralized Swarms" at Ninth International Conference on Bio-inspired Optimisation Methods and their Applications (BIOMA 2020) on Nov. 19, online.

Submitted on: Nov. 24, 2020

 

presented

David Mathias, Computer Science and Annie Wu, University of Central Florida presented "Dynamic Response Thresholds: Heterogeneous Ranges Allow Specialization While Mitigating Convergence to Sink States" at Twelfth International Conference on Swarm Intelligence (ANTS 2020 on Oct. 26 in Barcelona Spain (moved to online conference).

Submitted on: Nov. 24, 2020

 

presented

David Mathias, Computer Science; Laik Ruetten, CS: Embedded Sys BS; Annie Wu, University of Central Florida presented "Heterogeneous Response Intensity Ranges and Response Probability Improve Goal Achievement in Multi-agent Systems" at Twelfth International Conference on Swarm Intelligence (ANTS 2020) on Oct. 26 in Barcelona Spain (moved to online conference).

Submitted on: Nov. 24, 2020

 

published

David Mathias, Computer Science, co-authored the article "Evolutionary Optimization of Cooperative Strategies for the Iterated Prisoner’s Dilemma" in Transactions on Games and was accepted for publication by IEEE.

Submitted on: July 13, 2020

 

presented

Samantha Foley and David Mathias, both Computer Science, presented "A Parallel Two-stage Genetic Algorithm for Route Planning" at The Proceedings of the ACM Genetic and Evolutionary Computation Conference Companion (GECCO) on July 8 in Cancun, Mexico.

Submitted on: July 13, 2020