This is a picture of me at the Barrelman Triathlon in 2017. I used a cool technique called neural style transfer to “transfer” Monet’s style to the picture. (See implementation)

I am interested in diverse applications of Machine Learning and Data Science in health, policy, and social good. My expertise is in Hierarchical Multilevel Models and Computer Vision (Deep Learning – CNNs) to analyze brain images (fMRI/MRI). My research interests at the moment are meta-learning, few-shot learning, GAN’s in general, and biologically inspired models. I have a technical background in Computational Neuroscience, Machine Learning, and Physics.

These are some of the tools I’ve used in my work:

– Machine Learning (Reinforcement Learning, Deep Learning – Python/Scikit Learn/TensorFlow)
– Statistics (Multivariate, Bayesian – R)
– Neuroscience (fMRI, EEG analysis)
– Big Data (parallel processing, Spark, MapReduce)
– Decision Making (Behavioral Economics)

For my PhD at the University of Toronto, I used Reinforcement Learning and Hierarchical Multilevel Modelling on brain imaging data to understand how the brain supports decision-making.

Visit my Google scholar profile and my github to learn more.

Aside from my research engagements, I serve as Scientific Director for Potential Project International where I support trainers and consult on the science of mindfulness and mind training. I’m also the current president of the Machine Learning Club at the University of Toronto. In my spare time I practice mindfulness and loving-kindness meditation, compete regularly in triathlon races, and volunteer for Hospice Toronto.