MENU

You are here

Applications of Data Science to Natural Sciences

External Lecturer: 
Alessandro Treves
Carlo Baccigalupi
Stefano de Gironcoli
Mathew Diamond
Antonio Celani
Course Type: 
PhD Course
Academic Year: 
2021-2022
Period: 
first and second terms
Duration: 
18 h
Description: 

This module will present in a series of Masterclasses applications of data science to the cutting edge questions of research in various fields of Natural Sciences.

Alessandro Treves: Can information theory help understand information processing in our brain? (2h)

Carlo Baccigalupi: Data Science in Cosmological Observations (2h)

Stefano de Gironcoli: Machine Learning for Material Science (6h). Machine Learning approaches are gaining momentum in the field of atomistic simulation of materials. This short course will review the main approaches to generate interatomic potentials allowing large scale simulations with the accuracy of quantum mechanics at a fraction of the cost. Regression by Gaussian Processes, Neural Networks, Graph Neural Networks will be discussed.

Mathew Diamond: Computational neuroscience of perception (6h). This short module will present students some introductory notions about computational neuroscientific methodologies for approaching perception, memory, and decision making. The theme of the lessons is an experimental design that we call The Golden Triangle, as applied to behaviorally trained animals, involving a tightly controlled sensory input, the measurement of the perceptual and behavioral output of the animal, and measurement of the neuronal activity evoked by the sensory input. The sensory input and the simultaneous neuronal firing are linked by the method of sensory coding – the attempt to define the algorithms underlying the sensory-perceptual representation of the stimulus.

Antonio Celani: Introduction to Reinforcement Learning and applications (2h)

Research Group: 
Next Lectures: 

Sign in