Patricia Suriana

I am a PhD student in the Computer Science department at Stanford University, advised by Ron Dror. My reseach focuses on the application of machine learning to computational structural biology problem. Prior to Stanford, I worked as a software engineer at Google Research, where I work on Halide, an open-source DSL designed specifically for computational photography.

I did my MEng at MIT, where I was advised by Saman Amarasinghe. I've spent time at Facebook, MIT CSAIL, Microsoft, Square Enix, and Linear Technology. I did my bachelors also at MIT.

Email  /  CV  /  Biography  /  Thesis  /  Google Scholar  /  LinkedIn


I'm interested in application of machine learning to computational structural biology, programming language (domain specific language), distributed systems, search/path planning, and high-performance computing. Much of my research is about making things run faster.

TIRAMISU: A Polyhedral Compiler for Expressing Fast and Portable Code
Riyadh Baghdadi , Jessica Ray, Malek Ben Romdhane, Emanuele Del Sozzo, Abdurrahman Akkas, Yunming Zhang, Patricia Suriana, Shoaib Kamil, and Saman Amarasinghe
CGO, 2019

A polyhedral framework designed to generate high performance code for multiple platforms including multicores, GPUs, and distributed machines.

Parallel Associative Reductions in Halide
Patricia Suriana, Andrew Adams, Shoaib Kamil
CGO, 2017

A new Halide scheduling directive that permits parallelization or vectorization of Halide algorithms (reductions) which were previously inherently serial.

Cloned from here!