Statistical and Deep Learning to improve the desing of new materials for selective reaction of energetic interest
Mentor: Óscar Barquero Pérez
Phone: (+34) 914888463
University: Universidad Rey Juan Carlos
Partner Host Institution: IMDEA Energy - Photoactivated Processes (Researcher Víctor de la Peña) ( - 2nd year of applied research compulsory
This research line has been developed together with IMDEA Energy. The second year of applied research in IMDEA Energy is compulsory under this line.
Keywords: deep learning, sustainable fuels production, artificial photosynthesis

Statistical and Deep Learning to improve the desing of new materials for selective reaction of energetic interest

There is a huge interest on the study and design of new materials and their application on selective reactions of energetic interest for the sustainable fuels production and energy storage via artificial photosynthesis (CO2 reduction and water splitting). There is a vast body of literature about how to design and characterizations these new materials, both theoretical and experimental, which represent a huge labelled dataset. We propose to use statistical and deep learning techniques to learn from this data. The main goal is to be able to model and characterize the best procedures for designing new materials on reactions of energetic interests. Deep learning methods have been proven to be very powerful identifying fundamental characteristics in complex processes, namely the use of autoencoders. Eventually, the development of these deep learning models based strategies can be applied to other disciplines helping to design the new generation of functional materials with enhanced efficiency in many others energetic reactions.

Departament: Theory on Signal and Communication and Telematic Systems and Computing
Research Group:
More Information:
Relevants projects on the area: Programa Redes Eléctricas Inteligentes Comunidad de MADRID. IP: José Luis Rojo Álvarez. Ref: S2013/ICE-2933
Relevants publications on the area: 1.- Carlos Figuera, Óscar Barquero-Pérez, et al. (2014) Spectrally adapted Mercer kernels for support vector nonuniform interpolation, Signal Processing, 94, 421-423..
2.- Collado, L.; Reynal, A.; Fresno, F.; Barawi, M.; Escudero, C.; Perez-Dieste, V.; Coronado, J. M.; Serrano, D. P.; Durrant, J. R.; de la Peña O’Shea, V. A. “Unravelling the Effect of Charge Dynamics at the Plasmonic Metal/Semiconductor Interface for CO2 Photoreduction”. Nat. Commun. 2018, 9 (1), 4986