Interdisciplinary Seminar: How to build a social computer SS2025

Registration

Official registration page (LSF) for Computer Science students at the Saarland University: https://www.lsf.uni-saarland.de/qisserver/rds?state=verpublish&status=init&vmfile=no&publishid=158453&moduleCall=webInfo&publishConfFile=webInfo&publishSubDir=veranstaltung

Or send an email to: dimitra.tsovaltzi@dfki.de

Background

Computers, mobile phones, and wearables have become an integral part of our daily lives. The user interfaces are designed to interact with factual information, such as contextual data. This level of interaction may be adequate for certain applications. However, what if a computer system could understand users like humans do and adapt to their individual social situations to establish a social, human-like interaction? The focus is on social training systems and social companions, such as virtual communication, conflict resolution training, as well as synchronous support for well-being in different application settings: Robotics, job interviews, teacher training, health and care.

This interdisciplinary project-oriented seminar in Psychology and Artificial Intelligence explores the question of how to make computers social. It examines the concepts and theories that define being social, as well as how humans communicate socially and how emotions guide action in different strands of social life. The seminar also investigates how these concepts can be transferred into computer models and what is technically feasible. This study will present relevant concepts and theories, which will be discussed in the context of and applied to computer models and generative AI. As a proof of concept, interactive social computer applications will be created. The project will involve collaboration between students of Psychology and Computer Science, who will co-design, implement, and evaluate each application.

See previous edition for an example schedule: https://affective.dfki.de/teaching-2/social-computing-seminar/how-to-build-a-social-computer-ss2024/

Requirements

For Psychology students: knowledge in the areas of models of emotions, models of social interaction and requirements are appreciated. The seminar is open for last-semester students.

For Informatics students: knowledge in the areas of AI, HCI, and software design are appreciated. A bachelor degree in Computer Science (or equivalent) is preferable, but not an exclusion criterion.

The main language is English, but German is possible. Why? The seminar is about emotions and values that have to be discussed between psychology and computer science students. Students will get support in expressing themselves in one of the two languages, in which they feel more comfortable with.

During the seminar, a Teams Group for each project will be available to support exchange.

Project slides and short progress reports are in English!

Places: Psychology Students 9 Places / Computer Science Students 9 Places

Goal

In the seminar mixed student teams (computer science and psychology) will work on projects to design, implement and evaluate an affective application. There should be a lively exchange between the two disciplines to in order to create a well thought-out system.

Organisational information

The seminar will be DFKI in NB -1.63 (VIS-Room)
(You are very welcome to ask at the front desk for instructions on getting there)

Tuesdays between 14:00 and 16:00.

Starting date April 15.

Participation

In total, maximal 18 students can participate in the seminar. 9 psychology students and 9 computer science students.

Students of psychology obtain 4 ECTS for attending the seminar.

Students of computer science obtain 7 ECTS for attending the seminar. As usual for computer science, they will be graded. The grading includes the participation during the seminar, implementation and evaluation of the final project.

Structure

Besides the kickoff meeting, we will have weekly meetings (all participants, 90min) in which the seminar groups are presenting theirs projects. We will discuss your ideas, your project plans, concerns and localize problem solutions.

By end of June, each group will get 45 mins to present their implemented project in details.

Between the weekly meetings, you are expected to have individual group meetings to plan your progress presentations. Individual group meetings are meetings of the whole project team in a project team channel. As a team, you will find the best way to make sure you are ready to present your progress and that every member is contributing.

Schedule

#  Date  Seminar Goals  What to do?  
1  15.04  Kick Off Join Teams group   
2  22.04  General Material  Read assigned material 
Start looking at the assigned readings and propose a focus for your project. What are the open questions for you? (3-4 slides) Register in LSF! 
3  29.04  Project Topic Readings   Prepare project title and discussion of basic literature  
4  06.05  Pitch your idea   Discuss the idea – the basic research question for SoSci and CS  
5  13.05  Identify final tools    Discuss conceptualisation (tools, instruments, materials…)  
6  20.05  Finalize your idea  Discuss operationalisation   
7  27.05  Presentation I   Presentation I – Idea, conceptualization and operationalisation (15 mins per group) 
8  03.06  Progress   Propose a realisation of your idea 
9  10.06  Progress  Discuss the idea 
10  17.06  Progress Discuss the realisation and validation plan of your idea   
11  24.06  Progress Updates and discussion 
12  01.07  Progress Updates and discussion 
13  08.07  Progress  Discuss validation/results of the project  
14  15.07  Final Presentation   Executive presentation with limitations, conclusion, and future research/ Plan (30 mins per group) 

Projects

To be announced on the 2nd week. 

Material

Slides/Media 

How to give a great research talk (Youtube, Simon Peyton Jones, Microsoft Research Cambridge) 

General Literature

Software

  • VisualSceneMaker, an authoring tool for Virtual Characters. 
  • ALMA, a computational model of affect that implements the appraisal theory of emotions from the psychologists Ortony, Clore, and Collins. 
  • OpenSSI, a framework for analysing social signals in realtime. 
  • YALLAH, Yet another low‐level agent handler. 

Contacts

  • Dimitra (seminar advisor, EduTech, dimitra.tsovaltzi@mx.uni-saarland.de),    
  • Fabrizio (seminar advisor, Computer science, fabrizio.nunnari@dfki.de)