MRV

☕️

Martin Roa-Villescas

Docent / Ph.D. candidate

About Me

Martin Roa-Villescas earned his B.Sc. degree in Electronic Engineering from the National University of Colombia (UNAL) in Manizales, Colombia, in 2010. He received his M.Sc. degree in Embedded Systems from Eindhoven University of Technology (TU/e) in Eindhoven, The Netherlands, in 2013. Currently, he is pursuing a Ph.D. degree in Bayesian Machine Learning at TU/e. Between 2013 and 2018, he served as an embedded software designer at Philips Research in Eindhoven, The Netherlands. His research interests encompass probabilistic graphical models, open-source software, and robotics.

Download CV

Interests

Probabilistic inference
Open-source software
Linux
Robotics

Education

PhD in Probabilistic Graphical Models
Eindhoven University of Technology (TU/e)
M.Sc. in Embedded Systems
Eindhoven University of Technology (TU/e)
B.Sc in Electronic Engineering
Universidad Nacional de Colombia (UNAL)

📚 My Research

I am a research scientist specializing in probabilistic graphical models. My work focuses on optimizing the computational performance of exact inference in discrete models. Recently, I have begun exploring the development of robots using ROS2, with applications in the context of smart industry.

Please reach out to collaborate 😃

Featured Publications

Probabilistic Inference

Martin Roa-Villescas, Xuanzhao Gao, Sander Stuijk, Henk Corporaal, Jin-Guo Liu (2024). Probabilistic inference in the era of tensor networks and differential programming. In Phys. Rev. Res..

PDF Cite DOI

Patrick Wijnings, Martin Roa-Villescas, Sander Stuijk, Bert De Vries, Henk Corporaal (2024). On the Importance of the Execution Schedule for Bayesian Inference. In ACM TOPML.

Cite DOI

Martin Roa-Villescas, Jin-Guo Liu, Patrick Wijnings, Sander Stuijk, Henk Corporaal (2024). Pushing the Boundaries of Probabilistic Inference through Message Contraction Optimization. In Artificial Intelligence.

Cite DOI

Martin Roa-Villescas, Jin-Guo Liu (2023). TensorInference: A Julia package for tensor-based probabilistic inference. In JOSS.

PDF Cite Code DOI

Martin Roa-Villescas, Jin-Guo Liu, Patrick Wijnings, Sander Stuijk, Henk Corporaal (2023). Scaling Probabilistic Inference through Message Contraction Optimization. In CSCE'23.

Cite Code DOI

See all publications

Talks