Hi!

My name is Rose Cersonsky and I am a Postdoctoral Researcher in the Laboratory of Computational Science and Modelling (COSMO) at the Ecolé Polytechnique Fédérale de Lausanne (EPFL) in Lausanne, Switzerland.

My research is largely focused on computer-aided materials design. In my doctoral research, I ran simulations using statistical mechanics to understand the role that particle shape has in predicting colloidal self-assembly. This included a lot of geometry, physics, computer science, and mathematics (to my delight). I’ve also taken an interest into the field of photonic crystals, and how we can use crystallographic analysis to understand them.

At EPFL, I work with Prof. Michele Ceriotti to develop machine learning methods and approaches for atomistic simulations. Specifically, I focus on the role of materials featurization in machine learning models.

In my spare time, I conduct community outreach, run, travel, and search for delicious food.

Recent News:

For more news items, check out the news page!


Rose presented her work on Structure-property maps with Kernel Principal Covariates Regression and Improving Feature and Sample Selection with Principal Covariates Regression at the APS March Virtual Meeting . A recording of the talk is available online for the duration of the meeting.


Checkout Rose's latest pre-print, Improving Sample and Feature Selection with Principal Covariates Regression on the arXiv!


Rose's work contributing to Structure-property maps with Kernel Principal Covariates Regression is published in the Journal for Machine Learning: Science and Technology.


Rose's work contributing to Chemiscope, an interactive structure-property explorer, is published in the Journal of Open Source Software. The HTML version is available at chemiscope.org.


Rose's publishes an article in the Proceedings of the American Society for Engineering Education (ASEE) detailing the REACT Workshop entitled Bringing Science Education and Research together to REACT.