In a broad sense, I want to make models of human social behavior to improve our understanding of fields like Economics and Law. More specifically, I am interested in using adaptive evolutionary neural networks as computational models to test against the behavioral economics and behavioral psychology literature. I am currently working on adapting the Neuroevolution of Augmented Topologies (NEAT) algorithm developed at UT Austin by Ken Stanley and Risto Miikkulainen for use in computational models of human behavior. I am also interested, more generally, in the development of biologically plausible artificial neural network (ANN) models. Another strand of interest relates to data analysis. I am interested in combining evolution of neural network topologies and more traditional ANN methods for optimization of connection weights, such as backpropagation with stochastic gradient descent or simulated annealing that can run on parallel GPUs, for data analysis, with particular application to MRI and other brain-imaging data.
MIT Economics (’09); J.D., Boston University School of Law (’13).
© 2013 MIT Sloan Neuroeconomics Lab