Hello there!
Welcome to my personal page
I am a computer scientist at SRI International in Menlo Park. I am currently working on inducing 'fairness' in neural network training and reducing variance in off-policy policy estimation in reinforcement learning. I employ concepts from variational inference (disentanglement of the latent space) and information theory (mutual and conditional mutual information) in my day to day research. If you use Keras, please check out my github for a model parameter manipulation script that I'm sure you will find useful in your model pruning projects.
GithubRecent Papers
Posts
Some stuff I have been working on is listed below