russell j. hewett

assistant professor

mathematics & CMDA

Joining My Research Group

Pay attention to this space for research opportunities. I will post them here as they arise.

Postdocs: I may have openings for postdocs in 2021. Occasional funding for teaching postdocs may be available through the department. Contact me if interested.

Grad students: I currently have openings for MS and PhD students in the space of HPC for deep learning. Current VT MS and PhD students who are interested in working with me should contact me by e-mail to setup a meeting with me. Prospective graduate students should also contact me to alert me to their application to VT.

Undergrads: I occasionally have space for undergrad researchers. I encourage CMDA students who want to do research with me to do well in CMDA 3634, with me (preferred) or with another instructor.

Group News

  • 2021/05/27 – Russell J. Hewett was awarded Early Career Research Award from Department of Energy, Office of Science, Advanced Scientific Computing Research for Domain-Decomposition Induced Parallelism for Scientific Deep Learning at Extreme Scale.

  • 2021/04/26 – Russell J. Hewett, Ali Habibnia (VT economics), and Dawson Miller received an award from the Virginia Tech College of Science’s Lay Nam Chang Dean’s Discovery Fund for Deep Learning for Non-linear Common-factor Modeling and Forecasting in Economics.

  • 2020/12/04 – DistDL v0.3.1 released on GitHub. [Link].

computational machine learning & inverse problems group

group members


Russell J. Hewett

principle investigator


Dawson Miller

graduate student


Joe Weissman

graduate student


Daniel Hajialigol

undergraduate student


Xuanjie Chen

undergraduate student

group alumni


ThaoVy Nguyen

Data Scientist @ L3-Harris


Jacob Merizian

Entrepeneur @ QuickTech Medical


Thomas Grady

PhD Student @ Georgia Tech

research projects

Distributed Deep Learning

DistDL, Model parallelism, software, and more

High-frequency Helmholtz

3D sweeping pre-conditioners, parallelism, and more

Solar Data Analysis

Solar Data Analysis, often with Python and SunPy

Solar Tomography

Numerics and 3D reconstructions of the solar atmosphere

Subsurface Imaging

PySIT, wave propagation, FWI, and more

recent publications

We develop a suite of parallel primitives for high-order tensors in distributed deep learning.

In this study, we used high-frequency dissolved oxygen data and inverse modeling to estimate daily rates of GPP and R in an …

We present the first fast solver for the high-frequency Helmholtz equation that scales optimally in parallel, for a single right-hand …

DistDL is a Python package for model parallel deep learning in PyTorch.

recent & upcoming talks

Understanding the structure of Earth’s subsurface is critical in many science, engineering, and business domains, from earthquake …

Training deep neural networks (DNNs) in large-cluster computing environments is increasingly necessary, as networks grow in size and …

Training deep neural networks (DNNs) in large-cluster computing environments is increasingly necessary, as networks grow in size and …

While it is theoretically and computationally advantageous to pose the inverse problem of subsurface recovery in the frequency domain, …