About

I am a PhD student in the Complex Multiscale System group in the Department of Chemical Engineering at Imperial College London, supervised by Prof. Serafim Kalliadasis and Dr. Miguel Durán-Olivencia. I hold a President's PhD Scholarship. Prior to this, I worked as a lead cloud architect at Allfunds, and as a data scientist at Ebury. Earlier, I completed a double Master's degree in Industrial Engineering, and in Research in Engineering System Modelling at Comillas Pontifical University, ICAI, Madrid (Spain). At the same time I was finishing my Master's degrees, I collaborated as a research scholar at the Institute for Research in Technology (IIT), ICAI, under the supervision of Prof. Pablo Frías. There, I was involved in projects related to the mathematical modelling of electricity markets and power system operation, specially in the valuation of financial derivatives for electricity markets underpinned by machine learning models.

email: a.malpica-morales21@imperial.ac.uk

Research

My research focuses on the understanding and implementation of data-driven and machine learning techniques to model and predict complex systems. Applying these techniques to a variety of real complex systems (ranging from fluids, climate to social networks or financial markets) allows us to infer and extract the hidden dynamic laws that govern the time-evolution of such intricate systems. Obtaining these dynamic laws is essential to explain certain phenomena (varying spatiotemporal dynamics at different scales, chaos, phase transitions, etc.) that emerge as the result of the non-trivial interactions between the complex system’s constituents.

The leitmotif of my PhD is the study of complex systems in stationary and non-stationary conditions (in and out of equilibrium) combining physics and mathematics areas (stochastic processes, differential equations, statistical physics) with statistical learning frameworks (neural differential equations, kernel density estimation). The overarching objective is the development of a theoretical-computational framework that can be widely applied to represent empirically observed complex phenomena across different subject areas and application domains.