Evan Racah
ejracah at gmail dot com

I am now working for Waymo Research applying machine learning to the Waymo self-driving car!

I recently finished my master's degree at Mila working with Chris Pal on unsupervised visual representation learning. I did my Bachelor's at UC Davis. After graduating, I worked at the National Energy Research Scientific Computing Center at Berkeley Lab, where I worked on using deep learning to solve problems in climate and high energy physics, as well as scaling machine learning algorithms to supercomputing scale.

Google Scholar  /  Github  /  Twitter  /  CV

I'm interested in unsupervised visual representation learning, computer vision, reinforcement learning, and intuitive physics. My current research focuses on learning representations from sequential visual data, specifically representations that capture objects and other semantic state information.

A new method for unsupervised learning of state representations from visual RL environments, as well as a new benchmark for measuring these representations.

Supervise Thyself: Examining Self-Supervised Representations in Interactive Environments
Evan Racah, Christopher Pal
ICML Workshop on Self-Supervised Learning, 2019
arxiv / workshop version / code

An qualitative and quantitative examination of the features learned by several popular self-supervised methods in video-game-like environments

Deep Neural Networks for Physics Analysis on low-level whole-detector data at the LHC
Wahid Bhimji, Steven Andrew Farrell, Thorsten Kurth, Michela Paganini, Evan Racah
Journal of Physics: Conference Series, 2018
arxiv / code

Classifying RPV-Supersymmetry events with deep neural networks

ExtremeWeather: A large-scale climate dataset for semi-supervised detection, localization, and understanding of extreme weather events
Evan Racah, Christopher Beckham, Tegan Maharaj, Samira Ebrahimi Kahou, Mr. Prabhat, Chris Pal
NIPS (now NeurIPS), 2017
project page / code / arxiv

A high dimensional spatiotemporal dataset for detection of extreme weather events

Deep learning at 15pf: supervised and semi-supervised classification for scientific data
Thorsten Kurth, Jian Zhang, Nadathur Satish, Evan Racah, Ioannis Mitliagkas, Md Mostofa Ali Patwary, Tareq Malas, Narayanan Sundaram, Wahid Bhimji, Mikhail Smorkalov, Jack Deslippe, Mikhail Shiryaev, Srinivas Sridharan, Pradeep Dubey
Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis (Supercomputing), 2017
arxiv

Scaling Deep Learning for climate and HEP on a large supercomputer

Revealing Fundamental Physics from the Daya Bay Neutrino Experiment Using Deep Neural Networks
Evan Racah, Seyoon Ko, Peter Sadowski, Wahid Bhimji, Craig Tull, Sang-Yun Oh, Pierre Baldi, Prabhat
IEEE International Conference on Machine Learning and Applications (ICMLA), 2016
video demo / slides / code / arxiv

Categorizing high energy physics phenomena in an unsupervised way

Matrix factorizations at scale: A comparison of scientific data analytics in Spark and C+ MPI using three case studies
Alex Gittens, Aditya Devarakonda, Evan Racah, Michael Ringenburg, Lisa Gerhardt, Jey Kottalam, Jialin Liu, Kristyn Maschhoff, Shane Canon, Jatin Chhugani, Pramod Sharma, Jiyan Yang, James Demmel, Jim Harrell, Venkat Krishnamurthy, Michael W Mahoney
2016 IEEE International Conference on Big Data (Big Data), 2016
code / arxiv

Comparing Apache Spark with C and MPI for scientific analysis workloads

Application of deep convolutional neural networks for detecting extreme weather in climate datasets
Yunjie Liu, Evan Racah, Joaquin Correa, Amir Khosrowshahi, David Lavers, Kenneth Kunkel, Michael Wehner, William Collins
arXiv preprint arXiv:1605.01156, 2016

Classifying extreme weather events from simulation

H5spark: bridging the I/O gap between spark and scientific data formats on Hpc systems
Jialin Liu, Evan Racah, Quincey Koziol, Richard Shane Canon, Alex Gittens, Lisa Gerhardt, Suren Byna, Mike F. Ringenburg, Prabhat
Cray User Group, 2016
slides / code /

A plugin to Apache Spark to read in HDF5 Files

Legit website I used as a template