Skip to main content

Marta Sales Pardo

ICREA Acadèmia 2021

Universitat Rovira i Virgili · Experimental Sciences & Mathematics

Marta Sales Pardo

I am currently a Professor in the Department of Chemical Engineering at Universitat Rovira i Virgili and the co-director of the science and engineering of emergent systems lab (sees lab). I studied and obtained a PhD in physics in Univeristat de Barclona (2002). I preformed my postdoctoral work at Nortwhestern University with Luis Amaral. In 2008, I became a Research Assistant Professor in the same university, within the Northwestern Institute on Complex Systems, the NU Center for Clinical and Translational Science and the School of Engineering. In 2010, I became an Associate Professor at universitat Rovira i Virgili. Since 2019, I am in the advisory committee of the James McDonnell Foundation and. Since 2020, I am the Responsible for Equality of the School of Chemical Engineering. In 2021, I was named Fellow of the Network Science Society. Since July 2022 I am the president of Complexitat, the Catalan Association for the Study of complex Systems.

Research interests

Since 2009, we have lead the development of generative group-based models for complex networks amenable to Bayesian inference. Recently, we have developed Bayesian methods for model selection in a wide array of network types and contexts including: bioinformatics, molecular biology and physiology, ecology,  human decision-making, and science of science.

We have recently developed what we call a Bayesian machine scientist, an algorithm that can automatically obtain mathematical expressions from data by assessing their plausibility. The BMS solves caveats from current approaches to the symbolic regression problem and, by exploiting the deep connection between probability theory and statistical physics, can perform proper model averaging, thus opening another chapter in the ability of symbolic regression approaches to rival 'black box' methods in predicting unobserved data.


complex networks; complex systems; statistical inference

ICREA Memoir 2022