INQ0092522 - MACHINE LEARNING (MOD. B) (A) 2022-2023
Topic outline
-
INQ0092522 - MACHINE LEARNING (MOD. B) (A) 2022-2023 - LAST DIGIT ID 0-4 - PROF. FEDERICO CHIARIOTTI
Target skills and knowledge:
The aim of this course is to provide the fundamentals and basic principles of the learning problem as well as to introduce the most common machine learning algorithms. The course will be complemented by hands-on experience.
Examination methods:The evaluation of the acquired skills and knowledge will be performed using two contributions:
1. A written exam (30 points).
2. Three simple homeworks to be done using Python (up to 3 bonus points).Exam dates:
- 24/01/2023
- 09/02/2023
- 28/06/2023
- 07/09/2023
- 21/09/2023
Course Program
1. Motivation: components of the learning problem and applications of Machine Learning. Supervised and unsupervised learning.
2. Introduction: The supervised learning problem, Classes of models, Losses, Probabilistic models and assumptions on the data. Regression and Classification.
3. When is a model good? Model complexity, bias variance tradeoff/generalization (VC dimension, generalization error), Cross Validation.
4. Models for Regression: Linear Regression, linear-in-the-parameters models, regularization.5. Simple Models for Classification: Logistic Regression, Perceptron, Naïve Bayes Classifier
6. Kernel Methods: Support Vector Machines.
7. Random Forests
8. Neural Networks
9. Deep Learning: Convolutional Neural Networks, advanced models
10. Unsupervised learning: Cluster analysis, Linkage-based clustering, K-means Clustering.
11. Dimensionality reduction: Principal Component Analysis (PCA).Laboratories:
1. Introduction to Python
2. Linear models for regression and classification
3. Support Vector Machines
4. Neural Networks
5. Deep Learning with Keras
Textbooks:- Shalev-Shwartz, Shai; Ben-David, Shai, Understanding machine learning: From theory to algorithms, Cambridge University Press, 2014.
- Slides and other material will also be provided by the instructor.
-
Closed: Friday, 18 November 2022, 9:51 AM
-
Introductory lecture on September 28th:
- What is machine learning?
- Course fundamentals and basic information
- Logistics
The lecture will also be streamed on Zoom at the following link: https://aaudk.zoom.us/j/64958751533 -
This module is about the theory of machine learning: what is learnable, what guarantees we can get on the quality of the results and the quantity of data we need, and which kinds of problems can and cannot be solved by learning.
-
This module is about practical learning models: linear classifiers and regressors, SVMs, and related concepts.
-
This module is about neural networks and deep learning approaches.
-
This module examines the problem of learning from unlabeled data: in this case, we are not learning a function, but rather patterns in the distribution of the original data.
-
Link to the Keras introduction tutorial video: https://mediaspace.unipd.it/media/t/1_pdghjg3g
-
Opened: Wednesday, 16 November 2022, 12:00 AMDue: Tuesday, 29 November 2022, 11:59 PM
-
Opened: Tuesday, 29 November 2022, 12:00 AMDue: Tuesday, 13 December 2022, 11:59 PM
-
Opened: Wednesday, 14 December 2022, 12:00 AMDue: Tuesday, 10 January 2023, 11:59 PM
-
Closed: Friday, 16 December 2022, 4:01 PM