Conversational AI is a graduate level seminar intended to cover recent research progress on chat bots, task-oriented dialogue systems, conversational search and recommender systems, conversational question answering, multi-modal conversational systems, and related topics. Students are expected to read, present, and discuss research papers related to conversational AI systems. They will also get familiar with state-of-the-art tools and techniques for developing conversational systems.
Prerequisites: Either COMPSCI 682, COMPSCI 646, COMPSCI 685, or COMPSCI 670. Students are expected to be familiar with statistical, machine learning and deep learning models applied to natural language processing, information retrieval, and/or computer vision. 2 credits. Upon instructor’s permission, students can work on an optional project for 3 credits.
- Foundations of machine learning and deep learning models applied to natural language processing, information retrieval, and/or computer vision.
- [GGL] Jianfeng Gao, Michael Galley, Lihong Li. Neural Approaches to Conversational AI. Foundation and Trends in Information Retrieval, 2019.
- [ZTDR] Hamed Zamani, Johanne R. Trippas, Jeff Dalton, and Filip Radlinski. Conversational Information Seeking, 2022.
- Presentation (2CR: 50%, 3CR: 30%)
- Participation (2CR: 50%, 3CR: 30%)
- Final Project (2CR: 0%, 3CR: 40%)
- Bonus: Constructive feedback on ZTDR (2CR: 10%, 3CR: 10%)
|2||Background: Language Models||Tue 2/1||
Chit-Chat Conversation: History, Benchmarks, Evaluation Methodology
|4||Chit-Chat Conversation: Current State-of-the-Art and Future Directions||Tue 2/15||
|No Class (Monday Class Schedule)||Tue 2/22|
|5||Conversational Search: Retrieval, Benchmarks, and Evaluation||Tue 3/1||
|6||Conversational Search: Mixed-Initiative Interactions||Tue 3/8||
|No class: Spring break||Tue 3/15||
|7||Conversational Question Answering||Tue 3/22||
|8||Task-Oriented Dialogue Systems: History, Benchmarks, Evaluation Methodology||Tue 3/29|
|9||Task-Oriented Dialogue Systems: State-of-the-Art and Future Directions||Tue 4/5|
|10||Conversational Recommender Systems: Recommendation and Preference Elicitation||Tue 4/12||
|11||Multi-Modal Conversational Systems||Tue 4/19|
|12||Lessons Learned from Alexa Prize||Tue 4/26||
|13||Final Project Presentations||Tue 5/3|
Collaboration and Help
The work that you submit (e.g., Final Project) must be your own. It may not be copied from the web, from another student in the class, or from anyone else.