Course full name
Deep Learning for Physicists
Course ID number
2425-1-113R-03
Course Syllabus
Titolo
Introduction to Deep Learning for Physicists
Docente(i)
Stefano Giagu
Lingua
English
Breve descrizione
Monday June 16
- Lecture 1: Artificial Neural Networks 101
- Exercise session 1: introduction to the pytorch framework and implementation of a simple ANN
Tuesday June 17
- Lecture 2: Regularization and Training of an ANNs
- Exercise session 2: end-to-end design and training of an MLP to learn the dynamic of a mechanical system, Hamiltonian NNs
Wednesday June 18
- Lecture 3: Neural Network Models for Sparse Interactions: CNNs and Graph Neural Networks
- Exercise session 3: training a CNN to identify phase transitions on a 2D Ising Model
Thursday June 19
- Lecture 4: Unsupervised learning and Anomaly Detection with Auto Encoders
- Exercise session 4: representation of the latent space learnt by an AE, anomaly detection in high-energy physics
Friday June 20
- Lecture+exercise 5: Uncertainty quantification in ANNs, Bayesian NNs or recent trends in deep dearning
- Hackathon/final exam
Sustainable Development Goals
ISTRUZIONE DI QUALITÁ | IMPRESE, INNOVAZIONE E INFRASTRUTTURE
Title
Introduction to Deep Learning for Physicists
Teacher(s)
Stefano Giagu
Language
English
Short description
Monday June 16
- Lecture 1: Artificial Neural Networks 101
- Exercise session 1: introduction to the pytorch framework and implementation of a simple ANN
Tuesday June 17
- Lecture 2: Regularization and Training of an ANNs
- Exercise session 2: end-to-end design and training of an MLP to learn the dynamic of a mechanical system, Hamiltonian NNs
Wednesday June 18
- Lecture 3: Neural Network Models for Sparse Interactions: CNNs and Graph Neural Networks
- Exercise session 3: training a CNN to identify phase transitions on a 2D Ising Model
Thursday June 19
- Lecture 4: Unsupervised learning and Anomaly Detection with Auto Encoders
- Exercise session 4: representation of the latent space learnt by an AE, anomaly detection in high-energy physics
Friday June 20
- Lecture+exercise 5: Uncertainty quantification in ANNs, Bayesian NNs or recent trends in deep dearning
- Hackathon/final exam
Sustainable Development Goals
QUALITY EDUCATION | INDUSTRY, INNOVATION AND INFRASTRUCTURE