The project “Machine-Learning-Driven Atomistic Simulations for Energy and Biomedical Applications” will be led by the group of Modelling and Simulation in Life and Material Sciences at BCAM (Basque Country) and the MS2Discovery Interdisciplinary Research Institute at Wilfrid Laurier University (Waterloo, Canada). Both groups are involved in the International Consortium on Multiscale Modelling of Advanced Energy Materials and collaborate extensively with physicists, mathematicians, theoretical/experimental chemists and engineers from a number of institutions around the world.

The objective of the aforementioned project is to enable efficient and tractable simulations of several important classes of complex atomistic systems through the use of novel Machine Learning (ML) techniques, paying particular attention to those cases where state of the art Molecular Dynamics (MD) algorithms are lagging behind the current needs of challenging applications in energy and health. Many such applications require the modelling based on local atomic interactions for better understanding properties of the underlying structures and systems. Such interactions, often derived from quantum mechanical representations, are prohibitively expensive to simulate. ML algorithms provide a natural tool to address this challenge, and thus one of the underlying ideas of the project is to train neural networks efficiently or to use other ML tools in order to reproduce results of density functional theory calculations at a much lower cost.

The postdoctoral candidate will work under the supervision of Ikerbasque Research Professor Elena Akhmatskaya (MSLMS group, BCAM) and Roderick Melnik (Wilfrid Laurier University). Deadline: September 13th 2019, 14:00 CET. (UTC+1) .

Applications should be uploaded to the link:

http://www.bcamath.org/en/research/job/postdoctoral-fellowship-in-machine-learning-driven-atomistic-simulations-for-energy?utm_source=Recruitment&utm_campaign=7249757124-EMAIL_CAMPAIGN_2018_10_31_10_33_COPY_01&utm_medium=email&utm_term=0_d309d7bd58-7249757124-63770285