Affective Computing
A.Y. 2024/2025
Learning objectives
The course describes and analyses theory and techniques for the design of affective artificial agents that are able to:
- perceive the user's affective signals and extract their most significant cues
- infer, form such cues in a given context, user's affective state
- provide the appropriate feedback to the user
To such end, the course concerns: a rigorous introduction to the neurobiological and psychological models of emotions; stochastic processes and statistical machine learning and inference for modelling the dynamics of affect.
- perceive the user's affective signals and extract their most significant cues
- infer, form such cues in a given context, user's affective state
- provide the appropriate feedback to the user
To such end, the course concerns: a rigorous introduction to the neurobiological and psychological models of emotions; stochastic processes and statistical machine learning and inference for modelling the dynamics of affect.
Expected learning outcomes
Upon completion of the course students will be able to:
1. Define the methodology and the most appropriate techniques for modelling affective agents dealing with uncertainty
2. Measure and analyse affective signals, either behavioural or physiological and extract affective cues
3. Design and implement simple affective agents to be exploited in different applications such as video surveillance, autonomous driving, robotics, entertainment.
These objectives are measured via a combination of three components: the project realisation, the project technical report and the oral discussion. The final grade is formed by assessing the software developed, the project report, and then using the oral discussion for fine tuning.
1. Define the methodology and the most appropriate techniques for modelling affective agents dealing with uncertainty
2. Measure and analyse affective signals, either behavioural or physiological and extract affective cues
3. Design and implement simple affective agents to be exploited in different applications such as video surveillance, autonomous driving, robotics, entertainment.
These objectives are measured via a combination of three components: the project realisation, the project technical report and the oral discussion. The final grade is formed by assessing the software developed, the project report, and then using the oral discussion for fine tuning.
Lesson period: First semester
Assessment methods: Esame
Assessment result: voto verbalizzato in trentesimi
Single course
This course can be attended as a single course.
Course syllabus and organization
Single session
Responsible
Lesson period
First semester
Course syllabus
The rationale behind affective modelling addresses the following problems:
- detection and analysis of emotional signalling and display
- perception of the context in which signalling takes place
- understanding/inference of the hidden affective state behind signalling
Such problems involve both the neurobiological/psychological level of affect and the statistical computational level for designing models that suitably address the above issues.
The course is articulated in two parts:
Part 1 Neurobiological and psychological models of emotions
- Prelude to affective computing
- Emotion theories
- The computational landscape
- Folk theories of emotion
- The conceptual act theory
- In search of the emotional brain
- The predictive brain
- Other brains and language
Part 2 Probabilistic computational models of emotions
- Probabilistic models and the problem of Bayesian inference
- Approximate inference via Monte Carlo
- Approximate inference via Variational Bayes
- Probabilistic programming
- Deep generative neural networks and variational inference
- detection and analysis of emotional signalling and display
- perception of the context in which signalling takes place
- understanding/inference of the hidden affective state behind signalling
Such problems involve both the neurobiological/psychological level of affect and the statistical computational level for designing models that suitably address the above issues.
The course is articulated in two parts:
Part 1 Neurobiological and psychological models of emotions
- Prelude to affective computing
- Emotion theories
- The computational landscape
- Folk theories of emotion
- The conceptual act theory
- In search of the emotional brain
- The predictive brain
- Other brains and language
Part 2 Probabilistic computational models of emotions
- Probabilistic models and the problem of Bayesian inference
- Approximate inference via Monte Carlo
- Approximate inference via Variational Bayes
- Probabilistic programming
- Deep generative neural networks and variational inference
Prerequisites for admission
Students are encouraged to preliminarily attend the courses of Statistics and Statistical Machine Learning. Python programming capabilities are welcomed for project work.
Teaching methods
Lectures on general aspects of affective computing and, while developing the projects, hands-on activities within the PHUSE Lab
Teaching Resources
Lectures notes, live notes and slides, books, chapters, papers, videos and code are available on the course web site hosted by Ariel (https://gboccignoneac.ariel.ctu.unimi.it)
Assessment methods and Criteria
The examination concerns the development of a project and an oral discussion. The project is assigned based on ongoing affective computing research within the Perceptual computing and HUman SEnsing Laboratory (PHUSE Lab).
The project involves the realisation, and the experimental validation of a prototype addressing a specific case study. At the end of the project, the student should release a scientific report (illustrating the problem, results achieved, bibliography) and the code implementing the prototype.
The oral discussion aims at critically assessing project's motivation and results so far obtained, and the capability of the student to frame the latter within the theoretical topics presented in lectures.
The project involves the realisation, and the experimental validation of a prototype addressing a specific case study. At the end of the project, the student should release a scientific report (illustrating the problem, results achieved, bibliography) and the code implementing the prototype.
The oral discussion aims at critically assessing project's motivation and results so far obtained, and the capability of the student to frame the latter within the theoretical topics presented in lectures.
Educational website(s)
Professor(s)