Welcome to my homepage. I am currently a doctoral student at Indiana University in the School of Informatics and Computing.

My focus in computer science has been in artificial intelligence and machine learning. My dissertation research leverages techniques from these fields to model the evolution of communication. In particular, I am fascinated by the transition of humans from the primitive form of communication common in other animals into the richly powerful form of communication known as language. Language underlies much of our conscious activity and the nature of its emergence is an important open question in science.

My simulations involve populations of agents struggling to communicate their environment. Using Bayesian inference and techniques from information theory the evolution of these agents is guided by the principles of efficient communication, parsimony and fidelity. I am interested in whether agents develop communication that can be generalized to topics they haven't encountered before. Such generalization is the hallmark of a compositional language, one that is made up of components that can be arranged according to rules. The compositional nature of human language gives our communication an unlimited expressiveness that separates us from the narrow domains of communication in other animals.

Feel free to contact me at: gischrem AT cs.indiana.edu