Teaching plan for the course unit

 

 

Close imatge de maquetació

 

Print

 

General information

 

Course unit name: Probabilistic Graphical Models

Course unit code: 572671

Academic year: 2020-2021

Coordinator: Jerónimo Hernández González

Department: Department of Mathematics and Computer Science

Credits: 3

Single program: S

 

 

Estimated learning time

Total number of hours 75

 

Face-to-face and/or online activities

30

 

-  Lecture with practical component

Face-to-face and online

 

13

 

(Due to the Covid-19 restrictions, we expect to have 50% of in-person activities)

 

-  Problem-solving class

Face-to-face and online

 

6.5

 

-  IT-based class

Face-to-face and online

 

6.5

 

-  Student presentation and discussion

Online

 

4

Supervised project

15

Independent learning

30

 

 

Competences to be gained during study

 

CB6 ­ Poseer y comprender conocimientos que aporten una base u oportunidad de ser originales en el desarrollo y/o aplicación de ideas, a menudo en un contexto de investigación

CB9 ­ Que los estudiantes sepan comunicar sus conclusiones y los conocimientos y razones últimas que las sustentan a públicos especializados y no especializados de un modo claro y sin ambigüedades

CB10 ­ Que los estudiantes posean las habilidades de aprendizaje que les permitan continuar estudiando de un modo que habrá de ser en gran medida autodirigido o autónomo.

CE1 ­ Que los estudiantes sepan entender el proceso de valorización de los datos y su papel en la toma de decisiones.

CE2 ­ Que los estudiantes sepan recoger, extraer información y datos de fuentes de información estructuradas y no estructuradas.

CE7 ­ Que los estudiantes sepan entender, desarrollar y modificar los algoritmos analíticos y exploratorios que trabajan sobre conjuntos de datos y aplicar el pensamiento crítico en estas tareas.

 

 

 

 

Learning objectives

 

Referring to knowledge

To know what probabilistic graphical models (PGMs) are and what queries can we ask them.

 

 

To know when (and how) these queries can be answered exactly in polynomial time (exact inference).

 

To know what to do when they can not (approximate inference).
 

 

To know the basic techniques to learn probabilisitic graphical models from data.

 

Referring to abilities, skills

To be able to apply PGM algorithms to problems of your interest.

 

 

To be able to translate PGMs and related algorithms into code.

 

 

Teaching blocks

 

No..

Title

1

Representation

*  To understand what a probabilistic graphical model is and to interpret them in terms of statistical conditional independence assumptions and factorization.

2

Inference

*  To make probabilistic queries to a given PGM and answer them with exact and/or approximate algorithms.

3

Learning

*  To learn PGMs, both the structure and its parameters, from data.

4

Modern trends

*  To review state-of-the-art applications and tools in the area of PGMs.

 

 

Teaching methods and general organization

 

Due to the health situation by COVID19, teaching will be face-to-face (in-person), virtual (online) or mixed, following the instructions of the competent authorities. The most probable scenario for the 2020-21 academic year is a mixed teaching model.

* In case of in-person teaching:

Lectures dynamically combine master explanations and problem solving. Some slots may be exclusively dedicated to programming throughout directed activities or notebooks.

The students will be asked to make a presentation regarding an application (their own or other people’s) of PGMs to a problem of their interest, or a recently proposed study about PGMs.

* In case of mixed teaching (*expected scenario*):

In this case, we expect to a 50% of the activities to be in-person (about 1h, and the rest would be online). A methodology based on flipped classroom will be implemented:

- The in-person sessions will focus on problem solving and resolving practical doubts. If possible, evaluation activities will be carried out in person.

- The online part will be carried out synchronously and asynchronously. Theoretical master explanations will be posted so that students can access it asynchronously. Synchronous online sessions will be planned to ensure the correct development of the subject, with basic or support explanations and complementary problem solving activities.

* In case of online teaching:

The methodology of the previous mixed scenario are maintained, but the in-person teaching activities will be substituted by online synchronous sessions.

 

 

Official assessment of learning outcomes

 

The subject is expected to be evaluated based on a final exam (40%), a presentation (30%) and in-class activities (30%).

Depending on the health situation, the evaluable activities might be adapted to: face-to-face tests, synchronous online tests or work delivery.

 

Examination-based assessment

The subject is expected to be evaluated based on a final exam (50%) and a project (50%).