Deep Learning with Applications
A.Y. 2021/2022
Learning objectives
The course presents Deep Learning from a theoretical and practical point of view, introduces the basic elements of learning (non-linear models, minimization techniques, cross-validation and hyper-parameter tuning) and focuses on supervised, unsupervised and reinforcement learning models
Expected learning outcomes
At the end of the course the student will be able to:
- illustrate deep learning models in the learning context
supervised, unsupervised and for reinforcement.
- identify deep learning models suitable for the resolution of
problems in physics and beyond.
- use software and libraries for the development of deep learning models.
- illustrate deep learning models in the learning context
supervised, unsupervised and for reinforcement.
- identify deep learning models suitable for the resolution of
problems in physics and beyond.
- use software and libraries for the development of deep learning models.
Lesson period: Second semester
Assessment methods: Esame
Assessment result: voto verbalizzato in trentesimi
Single course
This course cannot be attended as a single course. Please check our list of single courses to find the ones available for enrolment.
Course syllabus and organization
Single session
Responsible
Lesson period
Second semester
Course syllabus
- Definition of Deep Learning, the current importance, challenges and limitations.
- Non-linear models, performance metrics and training techniques.
- Hyper-optimization techniques.
- Techniques for unbalanced training.
- Definition and examples of deep neural networks.
- Regression and classification models with deep learning.
- Generative models with deep learning.
- Reinforcement learning models with deep learning.
- Introduction to image recognition and object detection techniques.
- Short introduction to graph neural networks.
- Short introduction to Quantum Machine Learning models
- Non-linear models, performance metrics and training techniques.
- Hyper-optimization techniques.
- Techniques for unbalanced training.
- Definition and examples of deep neural networks.
- Regression and classification models with deep learning.
- Generative models with deep learning.
- Reinforcement learning models with deep learning.
- Introduction to image recognition and object detection techniques.
- Short introduction to graph neural networks.
- Short introduction to Quantum Machine Learning models
Prerequisites for admission
Basic knowledge in statistics, calculus and linear algebra.
Knowledge of at least one programming language.
Knowledge of at least one programming language.
Teaching methods
Lectures and exercises in the computational laboratory.
Teaching Resources
Textbooks:
- Deep learning, I. Goodfellow, Y. Bengio, A. Courville
- The elements of statistical learning, T. Hastie, R. Tibshirani, J. Friedman
- An introduction to statistical learning, G. James, D. Witten, T. Hastie, R. Tibshirani
Websites:
- Ariel: https://ariel.unimi.it/
- https://github.com/scarrazza/DL2022
- Deep learning, I. Goodfellow, Y. Bengio, A. Courville
- The elements of statistical learning, T. Hastie, R. Tibshirani, J. Friedman
- An introduction to statistical learning, G. James, D. Witten, T. Hastie, R. Tibshirani
Websites:
- Ariel: https://ariel.unimi.it/
- https://github.com/scarrazza/DL2022
Assessment methods and Criteria
Development of a project. The project topic has to be previously discussed with the lecturer. The project should demonstrate the comprehension of the lectures topics and the capability of proposing and motivating innovative solutions to specific research problems.
FIS/02 - THEORETICAL PHYSICS, MATHEMATICAL MODELS AND METHODS - University credits: 6
Lessons: 42 hours
Professor:
Carrazza Stefano
Professor(s)