Ján Drgoňa
Johns Hopkins associate professor and lead developer of PNNL's NeuroMANCER scientific machine learning library
- Associate Professor, Department of Civil and Systems Engineering (since January 2025), Johns Hopkins University12
Ján Drgoňa is an associate professor in the Department of Civil and Systems Engineering at Johns Hopkins University, whose work covers differentiable programming and scientific machine learning for dynamical systems, constrained optimization and control, with applications to sustainable energy systems. Drgoňa was previously a data scientist at Pacific Northwest National Laboratory and is a lead developer of NeuroMANCER, PNNL's open-source PyTorch library for constrained optimization, physics-informed machine learning and optimal control.123
Career
Drgoňa worked as a postdoctoral researcher at KU Leuven from 2017 to 2019, then joined Pacific Northwest National Laboratory as a postdoctoral researcher in July 2019 and became a data scientist there in 2020.2
As a Data Scientist 4 at PNNL from July 2023 to December 2024, Drgoňa led a team of research scientists working on differentiable programming methods for modeling and control of dynamical systems with energy applications. Drgoňa became an associate professor at Johns Hopkins University in January 2025 and holds a secondary appointment in electrical and computer engineering.21
Research and software
Drgoňa's research centers on differentiable predictive control, introduced in a 2020 paper with Aaron Tuor and Draguna Vrabie, and on physics-informed machine learning for dynamical systems, including a 2023 survey of physics-informed machine learning for modeling and control written with collaborators from PNNL and other institutions.451
NeuroMANCER, developed at PNNL, is a PyTorch library for solving constrained optimization, physics-informed machine learning and optimal control problems. Its README lists Drgoňa and Aaron Tuor as lead developers and Draguna Vrabie and Panos Stinis as scientific advisors.31
Work
NeuroMANCEROpen-source code
Listed with Aaron Tuor as a lead developer of NeuroMANCER in the repository's development team section, and first author of the library's recommended citation.31
Differentiable predictive control (external site: arxiv.org)Published research
First author, with Aaron Tuor and Draguna Vrabie, of the paper introducing differentiable predictive control (DPC), a method for offline learning of constrained neural control policies.41
Career timeline
Sep 2017
Started a postdoctoral position at KU Leuven.2
Jul 2019
Joined Pacific Northwest National Laboratory as a postdoctoral researcher.2
Apr 23, 2020
Posted the differentiable predictive control paper as first author.4
Jun 24, 2023
Co-authored a survey of physics-informed machine learning for modeling and control of dynamical systems.5
Jul 2023
Became a Data Scientist 4 at PNNL, leading a team on differentiable programming for dynamical systems.2
Jan 2025
Joined Johns Hopkins University as an associate professor.2
Other documented roles
Roles the sources document but that are not confirmed as current.
- Postdoctoral researcher and data scientist (2019 to 2024), Pacific Northwest National Laboratory2
Related people
More sourced profiles at the same organization, in the same field, or from the same school.
- Deb GracioAlso at Pacific Northwest National Laboratory
- Abdul Fatir AnsariAI for science
- Brandon DuderstadtAlso from Johns Hopkins
- Alexandre GramfortML frameworks and libraries
- George HotzAlso from Johns Hopkins
- Gilles LouppeAI for science
- Jared KaplanAlso from Johns Hopkins
- Guolin KeML frameworks and libraries
- Laura McGillAlso from Johns Hopkins
- Abhimanyu DasAI for science
- Margaret MitchellAlso from Johns Hopkins
- Abubakar AbidML frameworks and libraries
Is this you?
Profiles cover public professional work only, and every line cites a source. If something is out of date, missing or wrong, send a correction with a link to an official page and we will update it. Suggest a correction
Add this badge to your website or GitHub profile README. It links back to this page. It is not an endorsement or a rating.
Markdown (for a README)
[](https://unitedstatesofamericasuperintelligence.com/people/jan-drgona/)
HTML
<a href="https://unitedstatesofamericasuperintelligence.com/people/jan-drgona/"><img src="https://unitedstatesofamericasuperintelligence.com/badges/profile.svg" alt="Ján Drgoňa — Profile on USASI" height="20"></a>
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.