Human-centered AI: how can we support end-users to interact with AI?
0 likes38 views
The document summarizes research on developing human-centered AI systems that support end-users. It discusses explaining model outcomes to increase trust, enabling user interaction with explanations, and identifying what end-users need through evaluations. Strategies discussed include data-centric explanations, in-situ decision support, explaining model behavior, and addressing different needs in high-stakes domains like healthcare. The goal is developing explanations and interfaces tailored to non-experts through user-centered design and evaluations.
Human-centered AI: how can we support end-users to interact with AI?
- 1. Human-centered AI: how can we support end- users to interact with AI? Data Science Leuven – 10 May 2023 Katrien Verbert Augment/HCI – Department of Computer Science – KU Leuven @katrien_v https://augment.cs.kuleuven.be
- 3. Human-Computer Interaction group Explainable AI - recommender systems – visualization – intelligent user interfaces Augment Katrien Verbert ARIA Adalberto Simeone Computer Graphics Phil Dutré LIIR Sien Moens NLP Miryam de Lhoneux E-media Vero Vanden Abeele Luc Geurts
- 4. q Explaining model outcomes to increase user trust and acceptance q Enable users to interact with the explanation process to improve the model New forms of human-AI interactions Models
- 5. Explaining prediction models 5 Gutiérrez, F., Ochoa, X., Seipp, K., Broos, T., & Verbert, K. (2019). Benefits and trade-offs of different model representations in decision support systems for non-expert users. In Human-Computer Interaction–INTERACT 2019
- 6. Learning analytics & human resources Media consumption Wellbeing, food & health Precision agriculture FinTech & Insurtech What do end-users really need?
- 7. 7 Gutiérrez Hernández F., Seipp K., Ochoa X., Chiluiza K., De Laet T., Verbert K. (2018). LADA: A learning analytics dashboard for academic advising. Computers in Human Behavior, pp 1-13. doi: 10.1016/j.chb.2018.12.004 LADA: a learning analytics dashboard for study advisors
- 8. Study design 8 Evaluation @KU Leuven Monitoraat N = 12 6 Experts (4F, 2M) 6 Laymen (1F, 5M) Evaluation @ESPOL (Ecuador) N = 14 8 Experts (3F, 5M) 6 Laymen (6M)
- 9. Results 9 ✚ valuable tool for more accurate and efficient decision making. ✚ users evaluated significantly more scenarios. − more transparency needed increase trust. − model didn’t behave as expected − LADA didn’t meet our users needs Gutiérrez Hernández F., Seipp K., Ochoa X., Chiluiza K., De Laet T., Verbert K. (2018). LADA: A learning analytics dashboard for academic advising. Computers in Human Behavior, pp 1-13. doi: 10.1016/j.chb.2018.12.004
- 10. Strategies 10
- 11. Design science research 11 Fraefel, U. (2014, November). Professionalization of pre-service teachers through university-school partnerships. In Conference Proceedings of WERA Focal Meeting, Edinburgh.
- 12. Data-centric explanations Charleer, S., Moere, A. V., Klerkx, J., Verbert, K., & De Laet, T. (2018). Learning analytics dashboards to support adviser-student dialogue. IEEE Transactions on Learning Technologies, 11(3), 389-399.
- 13. Do not oversimplify: show uncertainty ¤ reality is complex ¤ measurement is limited ¤ individual circumstances ¤ need for nuance ¤ trigger reflection
- 14. 14 Charleer S., Gutiérrez Hernández F., Verbert K. (2019). Supporting job mediator and job seeker through an actionable dashboard. In: Proceedings of the 24th IUI conference on Intelligent User Interfaces Presented at the ACM IUI 2019 Actionable explanations
- 15. 15
- 17. Final evaluation 17 66 job seekers (age 33.9 ± 9.5, 18F) 8 Training Programs, 4 Groups, 1 Hour. 1 2 3 4 5 6 7 8 ResQue Questionnaire + two open questions. Users explored the tool freely. All interactions were logged.
- 18. 18 Results
- 19. 19 Results
- 20. Take away messages ¤ Explanations contribute to user empowerment ¤ Enable actionable insights ¤ Need for customization and contextualization ¤ Need for simplification 20
- 23. AHMoSe Rojo, D., Htun, N. N., Parra, D., De Croon, R., & Verbert, K. (2021). AHMoSe: A knowledge-based visual support system for selecting regression machine learning models. Computers and Electronics in Agriculture, 187, 106183.
- 24. AHMoSe Visual Encodings 24 Model Explanations (SHAP) Model + Knowledge Summary
- 28. 28
- 29. What if the stakes are really high? Learning analytics & human resources Media consumption health Precision agriculture FinTech & Insurtech
- 31. RECOMMENDER ALGORITHMS MACHINE LEARNING INTERACTIVE DASHBOARDS SMART ALERTS RICH CARE PLANS OPEN IoT ARCHITECTURE Explaining predictions health
- 32. 32 Gutiérrez Hernández, F. S., Htun, N. N., Vanden Abeele, V., De Croon, R., & Verbert, K. (2021). Explaining call recommendations in nursing homes: a user-centered design approach for interacting with knowledge-based health decision support systems. In Proceedings of the 27th Annual Conference on Intelligent User Interfaces. ACM. Explaining predictions health
- 33. Evaluation ¤ 12 nurses used the app for three months ¤ Data collection ¤ Interaction logs ¤ Resque questions ¤ Semi-structured interviews 33
- 34. ¤ 12 nurses during 3 months 34
- 35. Results ¤ Iterative design process identified several important features, such as the pending list, overview and the feedback shortcut to encourage feedback. ¤ Explanations seem to contribute well to better support the healthcare professionals. ¤ Results indicate a better understanding of the call notifications by being able to see the reasons of the calls. ¤ More trust in the recommendations and increased perceptions of transparency and control ¤ Interaction patterns indicate that users engaged well with the interface, although some users did not use all features to interact with the system. ¤ Need for further simplification and personalization. 35
- 36. 36
- 37. 37 Explaining recommendations Word cloud Feature importance Feature importance+ % Maxwell Szymanski, Vero Vanden Abeele and Katrien Verbert Explaining health recommendations to lay users: The dos and don’ts – Apex-IUI 2022 health
- 39. Results ¤ Hybrid explanations more useful compared to both the textual and visual explanations. ¤ Users with a higher NFC tend to score the hybrid explanations lower in terms of trust, transparency and usefulness compared to the unimodal explanation. 39
- 41. Combining XAI methods to address different dimensions of explainability ¤ Increasing actionability through interactive what-if analysis ¤ Explanations through actionable features instead of non- actionable features ¤ Color-coded visual indicators for easy identification of patients with high risk ¤ Data-centric directive explanations 41 Bhattacharya, A., Ooge, J., Stiglic, G., & Verbert, K. (2023, March). Directive Explanations for Monitoring the Risk of Diabetes Onset: Introducing Directive Data-Centric Explanations and Combinations to Support What-If Explorations. In Proceedings of the 28th International Conference on Intelligent User Interfaces (pp. 204-219).
- 42. Take-away messages ¤ Involvement of end-users has been key to come up with interfaces tailored to the needs of non-expert users ¤ Actionable vs non-actionable parameters ¤ Data-centric explanations provide powerful solution ¤ Need for personalisation and simplification 42
- 43. New research directions ¤ Conversational XAI methods ¤ Fair AI 43
- 45. Peter Brusilovsky Nava Tintarev Cristina Conati Denis Parra Collaborations Bart Knijnenburg Jurgen Ziegler