Luke Anderson

I am a PhD student at MIT CSAIL (lukea _at_ mit), where I am a member of the computer graphics group. My advisor is Frédo Durand.

My research interests include the design and implementation of domain specific languages for problems in computer graphics, and the application of machine learning to these languages and their compilers. I have contributed to 3 domain specific languages: Aether for Monte Carlo rendering, Halide for image processing, and Taichi for physical simulation. I'm interested in domain specific languages for their ability to make programming in a broad sense easier and this has been the primary goal of my work: Aether makes it easier to write complex rendering algorithms correctly, the Halide autoschedulers make it easier to find high performance implementations of image processing pipelines, and Taichi makes easier to write high performance code for spatially sparse computations, notably physical simulation.


Research

aether

Efficient Automatic Scheduling of Imaging and Vision Pipelines for the GPU

Luke Anderson, Andrew Adams, Karima Ma, Tzu-Mao Li, Tian Jin, Jonathan Ragan-Kelley

OOPSLA 2021

paper

difftaichi2020

DiffTaichi: Differentiable Programming for Physical Simulation

Yuanming Hu, Luke Anderson, Tzu-Mao Li, Qi Sun, Nathan Carr, Jonathan Ragan-Kelley, Frédo Durand

ICLR 2020

paper

taichi2019

Taichi: A Language for High-Performance Computation on Spatially Sparse Data Structures

Yuanming Hu, Tzu-Mao Li, Luke Anderson, Jonathan Ragan-Kelley, Frédo Durand

SIGGRAPH Asia 2019

paper

autoscheduler2019

Learning to Optimize Halide with Tree Search and Random Programs

Andrew Adams, Karima Ma, Luke Anderson, Riyadh Baghdadi, Tzu-Mao Li, Michael Gharbi, Benoit Steiner, Steven Johnson, Kayvon Fatahalian, Frédo Durand, Jonathan Ragan-Kelley

SIGGRAPH 2019

paper

aether2017

Aether: An Embedded Domain Specific Sampling Language for Monte Carlo Rendering

Luke Anderson, Tzu-Mao Li, Jaakko Lehtinen, Frédo Durand

SIGGRAPH 2017

paper













Accessibility