1st HEXED (The 1st Human-Centric eXplainable AI in Education) Workshop
The HEXED (Human-Centric eXplainable AI in Education) Workshop was held in conjunction with EDM (Educational Data Mining) 2024. The workshop aimed to bring together a specialized community of researchers who can work together to (1) develop a shared vision and common vocabulary for XAI in education, (2) share and disseminate work, (3) create robust methods for increasing interpretability, and (4) develop evaluation metrics for assessing explanations and model interpretability. We planned to achieve this through lively debate and discussion surrounding the current and future needs of the community.
The workshop was held on 14 July 2024 in Atlanta, Georgia, USA. This is a full-day hybrid workshop and will feature a mix of poster presentations, a lively panel discussion, and interactive sessions to facilitate collaboration.
Proceedings
The official joint proceedings (with the L3MNGET Workshop) can be found at CEUR-WS here.
Schedule
The workshop was held from 9:00am to 5:00pm.
The HEXED Workshop was held on Sunday, 14 July 2024 9:00AM–5:00PM UTC-4 (US-East). The schedule was as follows:
9:00–9:30am | Welcome and logistics |
9:30–10:30am | Poster session (+ lightning presentations) |
10:30–11:00am | Break + networking |
11:00am–12:00pm | Working session 1: Small group brainstorming |
12:00–1:00pm | Lunch break |
1:00–2:00pm | Keynote presentation: Personalized XAI (Cristina Conati) |
2:00–2:45pm | Working session 2: Framing problems and needs |
2:45–3:30pm | Panel |
3:30–4:00pm | Break + networking |
4:00–4:45pm | Working session 3: Creating a shared vision |
4:45–5:00pm | Closing thoughts |
Accepted papers
The official joint proceedings (with the L3MNGET Workshop) can be found at CEUR-WS here.
Research papers
- The Actionable Explanations for Student Success Prediction Models: A Benchmark Study on the Quality of Counterfactual Methods [paper @ CEUR-WS]
Mustafa Cavus and Jakub Kuzilek - Enhancing Explainability of Knowledge Learning Paths: Causal Knowledge Networks [paper @ CEUR-WS]
Yuang Wei, Yizhou Zhou, Yuan-Hao Jiang and Bo Jiang - Combining Cognitive and Generative AI for Self-Explanation in Interactive AI Agents [paper @ CEUR-WS]
Shalini Sushri, Rahul Dass, Rhea Basappa, Hong Lu and Ashok Goel
Position papers
- Towards a Unified Framework for Evaluating Explanations [paper @ CEUR-WS]
Juan Pinto and Luc Paquette
Encore papers
- Making Course Recommendation Explainable: A Knowledge Entity-Aware Model using Deep Learning [paper @ EDM proceedings]
Tianyuan Yang, Baofeng Ren, Boxuan Ma, Md Akib Zabed Khan, Tianjia He and Shin’Ichi Konomi - How Ready Are Generative Pre-trained Large Language Models for Explaining Bengali Grammatical Errors? [paper @ EDM proceedings]
Subhankar Maity, Aniket Deroy and Sudeshna Sarkar - Easing the Prediction of Student Dropout for everyone by integrating AutoML and Explainable Artificial Intelligence [paper @ EDM proceedings]
Pamela Buñay-Guisñan, Juan Alfonso Lara, Alberto Cano, Rebeca Cerezo and Cristóbal Romero - Evaluating the Explainers: Black-Box Explainable Machine Learning for Student Success Prediction in MOOCs [paper @ EDM proceedings]
Vinitra Swamy, Bahar Radmehr, Natasa Krco, Mirko Marras and Tanja Käser
Registration
The registration was managed by the EDM 2024 main conference organization at https://educationaldatamining.org/edm2024/. Registration is now closed.
Organizers
The workshop organizers are listed below. Click here to learn more about them or to see the full program committee.
- Juan D. Pinto, University of Illinois Urbana‐Champaign
- Luc Paquette, University of Illinois Urbana-Champaign
- Vinitra Swamy, EPFL
- Tanja Käser, EPFL
- Qianhui (Sophie) Liu, University of Illinois Urbana-Champaign
- Lea Cohausz, University of Mannheim