Multivariate analysis and Evolving processes
- Characterization, information, discrimination, prediction, Markov, …
Introduction to Machine Learning through supervised projects
- K-means, Kohonen neural networks, Generative algorithms, Machine learning, etc. Several topics (experimental and theoretical Chemistry, Molecular and Chemical Kinetics and Nanosciences) will be proposed to the students to apply the studied techniques.
Programming languages used in this course: Octave/Matlab and Python
Artificial Intelligence and Machine Learning are revolutionizing research and discovery in many scientific disciplines including Material Science, Nanotechnology and Pharmaceutical Chemistry. High throughput methods in material and pharmaceutical research often generate huge datasets that require data mining and Machine Learning techniques to extract relevant information that is needed to make new discoveries. Computational Physics and Chemistry are increasingly experimenting with Big Data techniques to extend and accelerate their approaches to larger and more realistic systems and simulations. Furthermore, Machine Learning methods may potentially solve the so-called Inverse Design Problem, the prediction of the structure of a novel material (or pharmaceutical compound) from its desired properties.
What do we offer?
This program has been conceived as a combination of knowledge and skill-based courses. It will provide the student with the fundamentals of a valuable and universal tool which can be adapted to multiple types of problems and situations in Molecular and Nanosciences. After an initial introduction on the theoretical methods necessary to understand the fundamentals of Machine Learning (ML), the student will be able to either choose the supervised ML project of his interest or conduct a ML-based research proposed by the instructors. This formation will be complemented by Keynote Lectures given by ML experts from technological companies and academia.
Upon completion of the course, the students should be able to:
- Manipulate scientific data and extract the relevant information
- Validate/invalidate theoretical hypotheses
- Characterize the past and the future of evolving processes.
- Apply Machine Learning to Molecular Sciences and Nanotechnology problems