Our research group focuses on studying and understanding 1) the underlying principles of biological computation, and how these principles can be adopted or modified to extend contemporary computer science methods, and 2) automated causal reasoning, such as abductive inference and Bayesian/belief networks.
Several properties of biologically-inspired computing separate it from more traditional computer science, giving hope that new robust and adaptive software methods can be developed. Examples of this type of computing include neural computation, evolutionary computation, artificial life, self-replicating machines, artificial immune systems, ant colony optimization, L-systems, artificial societies, and swarm intelligence. Our group has worked and/or is working in the following areas:
We are also focusing on automated causal reasoning using more traditional methods in artificial intelligence. The goal of this research is to model human cognition as a means of generating useful automated reasoning systems. Our group has worked and/or is working in the following areas:
We have an active seminar schedule that is open to interested individuals, and a recent undergraduate GEMSTONE research group working on genetic programming and multi-agent systems. Some simple online demonstrations illustrate a simulated self-assembling building , a neural model of cortical spreading depression, a a collection of cellular automata self-replicating structures, that were discovered using genetic programming, and a cellular automata model of self-replicating "machines" rendered with chess pieces.
We are studying methods for imitation learning by bimanual robots as part of an ONR-supported project.
We developed methods for self-organizing collective intelligence for adaptive problem solving as part of an NSF ITR project.
We also studied methods for computer-supported creativity enhancement using evolutionary computation as part of an NSF CreativeIT project.