We evaluated DARTS as an efficient way to discover neural architectures for materials science, astronomy, and medical imaging.

We distilled black-box reinforcement-learning policies into readable decision sets and measured the tradeoff between interpretability and performance.

Recently, in my role as Data Scientist at Amazon, I have been faced with one of the fundamental dichotomies of data science: the tradeoff between inference and prediction (aptly discussed in this H...

I recently re-read an article which I have become very fond of: “Artificial Intelligence: The Revolution Hasn’t Happened Yet” by one of the most respected researchers in the world of computer scien...

Last night, I started reading Malcolm Gladwell’s book, Outliers, and today I would like to write down my thoughts about it. In Outliers, Gladwell seeks to uncover the hidden alchemy of success and ...

We used sparse LSTM models and Granger-causal analysis to compare microbial interactions in healthy and IBD mice.

We trained bidirectional LSTM video classifiers over YouTube-8M using Spark, Elephas, HDFS, and an AWS EMR cluster.

We built a Python library for forward- and reverse-mode automatic differentiation, Jacobians, and gradient-based optimization.

We compared text representations and classifiers for identifying tweets from Internet Research Agency accounts.

We reproduced and evaluated a Bayesian GAN that uses stochastic-gradient MCMC to improve mode diversity and semi-supervised learning.