Current Courses

A few courses relating to the core interests of LCSL are offered at MIT and in Genoa. Those courses for which online resources are available, can be accessed by following the links below. Materials relating to these courses are constantly being updated. For the most recent information, contact the lecturers directly. Their info can be found on each course web page.

• RegML - Regularization Methods for Machine Learning
  Videos |  Description

  Taught by Lorenzo Rosasco. This is a one week course of roughly 11 lectures, including theoretical as well as practical (matlab) sessions. In many regards this is a short version of 9.520. It is held in the summer in Genova, Italy and is open to graduate students of any nationality. The course has steadily increased in attendance.
• TDII: Teoria dell'informazione e inferenza statistica (UniGE course)
  Description

  Taught by Alessandro Verri and Lorenzo Rosasco. The goal of this course is learning to extract information and infer knowledge from data via basic techniques from probability theory, statistics and information theory.
• MLCC - Machine Learning Crash Course
  Videos |  Description

  Taught by Lorenzo Rosasco. This is a week-long (crash) course including theoretical as well as practical sessions. MLCC usually takes place in Genova, Italy. The 2020 edition will take place in Oslo between the 25th and the 29th of May.
• Machine Learning (UniGE course)
  Description

  Taught by Lorenzo Rosasco and Nicoletta Noceti. This is a graduate term-long Machine Learning course offered at the University of Genoa.
• AML - Advanced Machine Learning (UniGE course)
  Description

  Taught by Annalisa Barla, Nicoletta Noceti and Lorenzo Rosasco. This is a graduate term-long course focussing on advanced topics such as representation learning, deep learning and learning in dynamic environments.
• MIT 9.520/6.860 - Statistical Learning Theory and Applications
  Videos |  Description

  Taught by Lorenzo Rosasco and Tomaso Poggio. This is a term long course of roughly 25 lectures offered to graduate students at MIT.
• Introduction to Convex Optimization
  Description

  PhD course taught by Saverio Salzo, Silvia Villa and Cesare Molinari. Convex optimization plays a key role in data sciences. The objective of this course is to provide basic tools and methods at the core of modern nonlinear convex optimization. Starting from the gradient descent method we will cover some state of the art algorithms, including proximal gradient methods and alternating direction method of multipliers, which are nowadays very popular techniques to solve machine learning and inverse problems.
• CBMM Summer School: Machine Learning Classes
  Videos |  Description

  Taught by Lorenzo Rosasco. This one day course provides an introduction to the essential concepts and algorithms at the core of modern machine learning. Theory classes in the morning are complemented by practical lab sessions in the afternoon.
• Mathematics for Machine Learning (UniGE course)
  Description

  Taught by Lorenzo Rosasco and Ernesto De Vito. TA: Nicolò Pagliana. The goal of this course is to provide the basic mathematical concepts for machine learning.
• Regularization: From Inverse Problems to Large Scale Machine Learning @ Summer School on Applied Harmonic Analysis and Big Data
  Description

  Taught by Lorenzo Rosasco.

Previous Courses

Courses organized by LCSL in previous years.

• VVV18 - International Winter School on Humanoid Robot Programming: Machine Learning Classes
  Description

  Taught by Lorenzo Rosasco. This short course provides an introduction to the essential concepts and algorithms at the core of modern machine learning (day 1) and deep learning (day 2). Theory classes in the morning are complemented by practical lab sessions in the afternoon.
• Spring School 'Structural Inference': Regularization Methods for Large Scale Machine Learning
  Description

  Taught by Lorenzo Rosasco.
• CBMM Summer School: Robotics Afternoon
  Description

  A series of five talks by leading researchers in robotics within the summer school organized by the Center for Brain Minds and Machines.
• Gaussian Process Winter School
  Description

  Taught by Neil Lawrence. Hosted by Lorenzo Rosasco in Genova, Italy.
• 9.s912 - Vision and learning - Computers and Brains
  Description

  Taught by Tomaso Poggio, Shimon Ullman and assisted by Ethan Meyers.

Independent Activity Period (IAP):

As part of the Independent Activities Period (IAP) at MIT, we have organized:

• The iCub-humanoid Robot: A Hands-on Course
  Description

  The course is organized as a 'code sprint' to pull together all the software needed to make a humanoid work - control and perception algorithms for its head, hand, arm, legs, and full body. It is organized as a peer-to-peer event: there are little formal lectures and lot of free programming time on a real humanoid robot. The bulk of the time will be spent working together to integrate modules and implement interesting behaviors for humanoid robots. A full iCub is available to students for the duration of the course (5 days). The robot is equipped with cameras, microphones, encoders, force/torque sensors, sensorized skin, gyros, accelerometers and controls 53 motors.
• The Workshop: Robotics and Learning Theory at the Istituto Italiano di Tecnologia
  Description

  As we started to work on applications of machine learning in robotics the workshop is an occasion to hear about the state of the art and the connections in the two fields.