Dr. Terry Soule
Dr. Terence Soule is a Professor of Computer Science at the University of Idaho
Part of the project I'm most interested in
I would like to see the genome made more accurate: multiple alleles per trait, diplody, etc. Then it will be possible to put in more interesting reproductive strategies, starting with sexual reproduction, but including mate selection, which could lead to nest building, courtship displays, etc. This is important because richer genetics and reproduction leads to richer ``results. E.g. the evolution of flowering plants created a huge burst of evolution, including much richer species diversity, much wider ranges of (beautiful) morphologies, co-evolution with pollinators that created changes in other kingdoms (butterflies and humming birds), etc.
I would like to work on more diverse environments. Again this should create more diversity. There are some fairly strong theoretical arguments that evolution by itself cannot generate complexity, rather that the complexity is a result of organisms evolving to survive in a complex world. Thus, the more complex the environment, different terrains, winds, maybe day/night or seasonal cycles, etc. the more complex creatures we are likely to see. Plus, it would look cool and starting with different random environments would hugely increase replayability.
My research is in evolutionary computation, a field of machine learning in which the learning process is modeled on evolution. I.e. computers evolving to be smarter. This includes work with evolutionary robots, evolutionary games, and understanding the evolutionary process as it applies to silicon and carbon based agents.
Everything from introductory programming for non-majors (and coding camps for middle school students) through graduate courses on evolutionary robotics.
The early Total War games, Civilization, Kingdom Rush - games that require strategy and planning. Also some role-playing games like Diablo and Neverwinter Nights. But, writing a game is more fun than playing one :)