Research in Social Signal Processing and Affective Computing

Social Signal Processing and Affective Computing (SSPAC) is a research field aimed at developing computational tools that can support human social and affective experience through machine analysis and production of multimodal nonverbal and verbal behavior. As AI Technology becomes more advanced, the relationship of Technology with Humans is becoming ever closer. There is a significant risk of designing technology without understanding the fundamental nature of the human condition and how this governs how we interact with others, react to situations, and learn from each other.

In this course you will gain an advanced understanding of how state of the art computational approaches can exploit social and affective signals. Accomplishing this requires an interdisciplinary approach that is grounded in theory and practice from both Data Science and AI Technology as well as Social, Cognitive and Affective Science. As a result this course provides you with the opportunity to combine the most up to date theory in Social, Cognitive, and Affective Science with cutting edge Machine Learning and AI Techniques (e.g. Foundation Models, Deep Learning, Reinforcement Learning, etc)

The main component of this course involves a course project. You will gain first hand experience through a group project where you will be required to design, implement, and analyse an entire research and development pipeline on this topic. Through this learning activity, you will develop advanced skills on developing and motivating a research question, design and implementation of an experiment and communication of these research results.

The first portion of the course involves a crash course in Social Signal Processing and Affective Computer. For students who have already taken the DSAIT HCAI Theme, this will be a refresher and also opportunity to deepen knowledge through more reading across topics related to the field and also with new topics. For students who have not taken this variant before, be prepared for a steep ( and hopefully rewarding!) learning curve.