(jump to teaching section)
Brian J. Ross
Dept of Computer Science
1812 Sir Isaac Brock Way
St Catharines, Ontario, Canada L2S 3A1
Office hours: --
Bio-Inspired Computational Intelligence Group. (BICIG)
My research area is computational intelligence, evolutionary computation,
and genetic programming. My students and I are active in evolutionary design research,
and the new fields of computational aesthetics and creativity.
One research project explored the evolution of image filters, which attempt to
duplicate a target colour palette.
We incorporated Dr Bill Ralph's mathematical model of aesthetics, with the
goal of evolving visually pleasing images.
We extended this research to incorporate ideas from work in non-photorealistic
We also explored evolutionary design and 3D modeling, where topics include
architecture, floor plan design, illumination of interior spaces,
generalized 3D model generation using aesthetic evaluation, and energy efficiency.
Other research has used genetic programming to evolve communicating agents for game
domains, shape characterization during image evolution, procedural textures for 3D surfaces,
and vector graphics images.
I am also interested in the automatic synthesis of bio-networks
encoded in stochastic process algebra, as well as higher-level
bio-network modeling languages such as logical gene gates and PIM.
The goal is to automatically synthesize bio-networks that could generate
given time-course data, for example, changing protein levels over time.
This research needs to consider the evaluation of often noisy time-course data, which is
best characterized by statistical analyses.
The research also makes use of high-dimensional multi-objective strategies, as well
as grammatical modeling of target languages for genetic programming.
Other research interests in computational intelligence include multi-objective and
many-objective optimization, feature reduction, GPU-acceleration of genetic programming,
classification, and virtual photography.
To support my research, I developed a Prolog-based genetic programming
system called DCTG-GP (Definite Clause Translation Grammar for Genetic
Programming). DCTG-GP lets the user define their target
language using a logical CFG.
This environment permits the languages grammar, semantics and constraints to be
Please see the BICIG web page for more information.
Arpi Sen Gupta
(jump to teaching section)
Biomodeling and Genetic Programming
Image Evolution Using 2D Power Spectra
Non-Photorealistic Rendering with Cartesian Genetic Programming using GPUs
Evolved Communication Strategies and Emergent Behaviour
of Multi-Agents in Pursuit Domains
Statistical Image Analysis for Image Evolution
Feature Selection Using Age-Layered Genetic Programming
Interior Illumination Design Using Genetic Programming
Online Texture Classification Using GPU-based Genetic Programming
Passive Solar Building Design Using Genetic Programming
Genetic Programming for Non-Photorealistic Rendering
Particle Swarms and Aesthetic Virtual Photography
Enabling and Measuring Complexity in Evolved Architecture
Aesthetic 3D Model Evolution
Evolution of Floor Plans
Evolving Conceptual Building Architectures
Evolving 2D vectorized images
Evolving 2D image filters
Evolving 2D textures using a model of aesthetics
Evolving 3D procedural textures
DCTG-GP: Logic grammars and GP
Useful links for students
Very best Whistler Resort Web Cams ♦ Whistler Blackcomb Mountain Cams
Department of Computer Science