Course Syllabus

Export

Il corso fornirà i principali concetti e strumenti operativi, basati su metodi computazionali, per rappresentare il processo di apprendimento e le tecniche di ragionamento in condizioni di incertezza. Gli studenti acquisiranno abilità nell'utilizzare i concetti e i metodi appresi per risolvere problemi decisionali. In particolare gli studenti acquisiranno le seguenti competenze: identificazione delle relazioni tra parametri usando modelli probabilistici, costruzione di modelli decisionali, identificazione e valutazione del modello decisionale.

Rappresentazione dell’incertezza nei problemi di decisione

Rappresentazione della conoscenza in ambienti incerti

Reti Bayesiane Incertezza e scelte razionali

Il ragionamento probabilistico nel tempo

Inferenza nei modelli dinamici

  1. "Representing uncertainty in decision problems Basic notions of probability theory Bayes rule and its application". Chapter 13.

2.1 "Knowledge representation in an uncertain domain Bayesian network semantics; Efficient representation of conditional probabilities". Chapter 14 (14.1, 14.2, 14.3).

2.2 D-separation (materiale fornito dal docente)

2.3 Generazione numeri psudo-casuali per campionamento (materiale fornito dal docente)

  1. "Exact and approximate inference in Bayesian Networks". Chapter 14 (14.4, 14.5)

  2. "Markov Chains" (materiale fornito dal docente)

  3. Hidden Markov Models; Forecasting, Filtering and Smoothing ". Chapter 15 (15.1, 15.2 15.3).

Nozioni di base di: probabilità, statistica, algebra lineare

Lezioni, esercitazioni in aula, laboratorio

Il corso è erogato in lingua italiana.

S. Russel, P. Norvig. “Intelligenza Artificiale: Un Approccio Moderno”, Prentice Hall, III Edizione

papers & slides

Secondo Semestre

Esame Scritto + orale facoltativo

Su appuntamento

Export

The course will provide the main concepts and operative tools, based on computational methods, for representing the learning process and the reasoning techniques in uncertain domains. Students will gain the ability of using the concepts and methods learned for solving practical operational decision problems. In particular, they will acquire the following abilities: to identify relations between parameters by using probabilistic models, to build models for decision making, to evaluate and find the problem solutions.

Representing uncertainty in decision problems

Knowledge representation in uncertain domains

Bayesian Networks

Pseudo-number generation for sampling

Inference on BN

Probabilistic Reasoning over time

Markov Chains

Hidden Markov Models

Inference in dynamic models

  1. "Representing uncertainty in decision problems Basic notions of probability theory Bayes rule and its application". Chapter 13.

    2.1 "Knowledge representation in an uncertain domain Bayesian network semantics; Efficient representation of conditional probabilities". Chapter 14 (14.1, 14.2, 14.3).

    2.2 D-separation (papers & slides)

    2.3 Pseudo-number generation for sampling (papers & slides)

    1. "Exact and approximate inference in Bayesian Networks". Chapter 14 (14.4, 14.5)

    2. "Markov Chains" (papers & slides)

    3. Hidden Markov Models; Forecasting, Filtering and Smoothing ". Chapter 15 (15.1, 15.2 15.3).

Basic notions of: probability, statistics, linear algebra

The course is in Italian.

Lectures, classroom exercises, lab exercises

S. Russel, P. Norvig. “Artificial Intelligence: A Modern Approach", Prentice Hall, III Edizione

papers & slides

Second Semester

Written Exam + oral (optional)

By appointment

Enter

Key information

Field of research
MAT/09
ECTS
6
Term
Second semester
Activity type
Mandatory to be chosen
Course Length (Hours)
52
Degree Course Type
2-year Master Degreee
Language
Italian

Staff

    Teacher

  • EF
    Elisabetta Fersini
  • GR
    Giulia Rizzi

Students' opinion

View previous A.Y. opinion

Bibliography

Find the books for this course in the Library

Enrolment methods

Manual enrolments
Self enrolment (Student)