Human-Machine Dialogue
Master in Artificial Intelligence Systems
Register to the Moodle page of this course to receive updated information regarding assignments, notices,
and general information.
Description
Robots capable of engaging in meaningful conversations with humans are becoming increasingly prevalent in various industries and consumer sectors. In this course, we delve into the fundamental principles of human-computer interaction and then examine the underlying structures of human dialogue. We draw from conversation linguistics to develop computational dialogue models, dialogue system architectures, and their evaluation. The second part of the course equips students with methodologies for designing conversational agents, data-driven training (including generative AI models), design tools, and a project-based lab that tackles real-world use cases. This practical approach ensures that you understand the concepts and learn how to apply them in real-life scenarios.
Students will design, develop, and evaluate human-machine dialogue systems for an application with large language models and agents.
Topics include core components, design methodologies, and recent advances such as retrieval-augmented generation and agentic frameworks.
=======Lecture Topics, Slides, Lab============
Below are the topics that the course will cover. The course is project-based, meaning
that the student is required to design, train, and develop a human-machine dialogue system by
the end of the course, whether the "touch point" is a smartphone, desktop web browser, a smart speaker,
a smartwatch, in-car "invisible" interface. The course is based on lectures, lab sessions, and project coaching.
The material gets updated in the course of the semester every week.
Course Description (Incl. Assignments and Student Evaluation)
Basics of Human-Computer Interaction
Linguistics of Conversations
Conversational Design and Wireframing
Emotions in Dialogue
Dialogue Models
Natural Language Generation
Dialogue Evaluation
Reinforcement Learning and Applications to HMD
Large Language Models ( Lecture and Lab)
Privacy for Data Collection
Crowdsourcing for Data Collection
Ethics and Conversational AI
Lab Workplan for the HMD project
Project Reports (from top-grade students) : Spoken HMD with Alexa, Project with LLMs
