Partners
ModML is made possible by the SLING Project, funded by a Consolidator ERC Grant, and by the contributions of the DIBRIS and DIMA Departments of UniGe.
At a glance
ModML is a Summer School on the latest developments in Machine Learning, open to students from any University upon selection. The course objective is to have experts presenting advanced topics in an introductory and insightful manner, allowing students to further dig in as they see fit.
The school will take place exclusively in person, it will not be streamed online.
-
This is an advanced machine learning course covering some of the topics of interest in modern machine learning. After a boot camp on the first day (on statistical learning, machine learning models and optimization for machine learning), the rest of the week will be dedicated to introducing modern topics, including: implicit regularization, sketching, reinforcement learning, machine learning for inverse problems, optimal transport for machine learning, fairness in machine learning.
-
Each presentation, held in the morning, will be introductory and self-contained, with an associated practical session in the afternoon. The last day will be dedicated to a workshop with invited speakers.
-
The course is advanced and assumes students to have already taken an introductory ML course. Similarly, basic programming skills, probability and linear algebra/calculus knowledge is assumed. More advanced tools will be introduced in class in a self-contained manner.
Logistics
The school will be held in our beautiful Genova, in the DIBRIS/DIMA building in Via Dodecaneso 35, 16146, home of MaLGa Center.
-
The course will be held in the DIBRIS Department building of Via Dodecaneso 35, Genova. Indications to the classroom will be visible at the entrance. You will have access to vending machines during the breaks betweein classes. We suggest bringing your own laptop, but please keep into consideration that, due to the number of expected participants, your battery should be fully charged.
We will have a 1.5-hour lunch break every day: although there is a bar close to the department, you are encouraged to bring your own lunch and enjoy the break with your peers. There are common areas just out of the classroom where you will be able to sit and have lunch.
Speakers & instructors
Classes will be held by members of the SLING Project team and MaLGa Center faculty members from UniGe’s DIBRIS and DIMA Departments.
Programme
Each presentation, held in the morning, will be introductory and self-contained, with an associated practical session in the afternoon, held by MaLGa researchers. The last day will be dedicated to a workshop with invited speakers.
Monday, June 19, 2023
Class
14:00-15:30 - Statistical Learning Theory
Lorenzo Rosasco
Class
16:00-17:30 - Optimization for Machine Learning
Silvia Villa
Tuesday, June 20, 2023
Class
09:30-11:00 - Sketching
Antoine Chatalic
Class
11:30-13:00 - Implicit Regularization
Cesare Molinari
Lab
14:15-15:45 - Lab - Sketching
Lab
16:15-17:45 - Lab - Implicit Regularization
Wednesday, June 21, 2023
Class
09:30-11:00 - Reinforcement Learning
Agnese Seminara
Class
11:30-13:00 - Machine Learning for Inverse Problems
Giovanni S. Alberti
Lab
14:15-15:45 - Lab - Reinforcement Learning
Lab
16:15-17:45 - Lab - Machine Learning for Inverse Problems
Thursday, June 22, 2023
Class
09:30-11:00 - Optimal Transport for Machine Learning
Simone Di Marino
Class
11:30-13:00 - Fairness
Nicolas Schreuder
Lab
14:15-15:45 - Lab - Optimal Transport for Machine Learning
Lab
16:15-17:45 - Lab - Fairness
Friday, June 23, 2023
Workshop
09:30-11:00 - Learning in interpolation regimes
Claire Boyer
Coffee break
11:00-11:30 - Coffee break offered by ERC-Sling
Workshop
11:30-13:00 - Sampling as first-order optimization over a space of measures
Anna Korba
Lunch
13:00-14:30 - Lunch offered by ERC-Sling
How to apply
Applications are open until midnight on the 16th of April; notifications of acceptance will be sent by the 23rd of April.
Application
Sunday, April 16, 2023
Registration Fees
Upon acceptance of your application, you will be requested to pay a registration fee, depending on your role and affiliation.
| Role | Fee |
|---|---|
| UniGe and IIT affiliates | waived |
| Non-UniGe students and postdocs | €50 |
| Non-UniGe professors | €100 |
| Professionals | €300 |
Organization
An advanced machine learning course
Only for information which is not available online, please refer to MaLGa’s Lab Manager.