Statistical methods for machine learning

A.A. 2019/2020
Insegnamento per
6
Crediti massimi
48
Ore totali
SSD
INF/01
Lingua
Inglese
Obiettivi formativi
Upon completion of the course students will be able to:
1. understand the notion of overfitting and its role in controlling the statistical risk
2. describe some of the most important machine learning algorithms and explain how they avoid overfitting
3. run machine learning experiments using the correct statistical methodology
These objectives are measured via a combination of two components: the project report and the oral discussion. The final grade is formed by assessing the project report, and then using the oral discussion for fine tuning.

Struttura insegnamento e programma

Edizione attiva
INF/01 - INFORMATICA - CFU: 6
Lezioni: 48 ore
STUDENTI FREQUENTANTI
Programma
Introduction
The Nearest Neighbour algorithm
Tree predictors
Statistical learning
Cross validation
Risk analysis of Nearest Neighbour
Risk analysis of tree predictors
Consistency and nonparametric algorithms
Linear classification
Online gradient descent
From sequential risk to statistical risk
Kernel functions
Support Vector Machines
Stability bounds and risk control in SVM
Boosting
Compression bounds
Neural networks and deep learning
Prerequisiti e modalità di esame
Prerequisites: calculus, probability and statistics, linear algebra.

The exam consists of:
- an experimental project or theory project to be summarized in a written report
- an oral discussion of the report including questions on a choice of topics covered in the course.
Metodi didattici
Frontal teaching
Guest lecturers from industry
Materiale didattico e bibliografia
Handouts
Shai Shalev-Shwartz e Shai Ben-David, Understanding Machine Learning: From Theory to Algorithms, Cambridge University Press, 2014.
STUDENTI NON FREQUENTANTI
Programma
Introduction
The Nearest Neighbour algorithm
Tree predictors
Statistical learning
Cross validation
Risk analysis of Nearest Neighbour
Risk analysis of tree predictors
Consistency and nonparametric algorithms
Linear classification
Online gradient descent
From sequential risk to statistical risk
Kernel functions
Support Vector Machines
Stability bounds and risk control in SVM
Boosting
Compression bounds
Neural networks and deep learning
Materiale didattico e bibliografia
Handouts
Shai Shalev-Shwartz e Shai Ben-David, Understanding Machine Learning: From Theory to Algorithms, Cambridge University Press, 2014.
Periodo
Secondo semestre
Periodo
Secondo semestre
Modalità di valutazione
Esame
Giudizio di valutazione
voto verbalizzato in trentesimi
Docente/i
Ricevimento:
Mercoledì 9:30-12:30
via Celoria 18, stanza 7007