education as a computational science
play

Education as a computational science Pierre Dillenbourg, EPFL EPFL - PowerPoint PPT Presentation

Education as a computational science Pierre Dillenbourg, EPFL EPFL MOOCS: 1908876 Hype is over but MOOCs continue to grow . 1815471 Registrations 77 Courses Online 97510 Passed 35 Courses In Preparation N=1728 0.4


  1. Education as a computational science Pierre Dillenbourg, EPFL

  2. EPFL MOOCS: 1’908’876

  3. Hype is over but MOOCs continue to grow …. 1’815’471 Registrations 77 Courses Online 97‘510 Passed 35 Courses In Preparation

  4. ● N=1728 0.4 ● N=938 EPFL Freshmen ● N=5295 0.2 EPFL Grade Baccalaureat Level ● HI LO N=1334 0.0 N=951 − 0.2 N=5666 NONE VIEW ACT MOOC Usage Patrick Jermann, Francisco Pinto (EPFL CEDE)

  5. Education Technologies Learning Analytics Teaching CS CS ED

  6. e-learning ????????? Pierre Dillenbourg, EPFL

  7. Swarm Cellulo (Ayberk Ozgur, Wafa Johal, P. DIllenbourg)

  8. Swarm Interactions ED CS Education Technologies

  9. Johal, Lemaignan, Asselborn, Jacq, Billard, Paiva, Dillenbourg

  10. Swarm Interactions Teachable Agents ED CS Education Technologies

  11. Education Technologies Learning Analytics Teaching CS CS ED

  12. learning Analytics predict, classify, decide, ‘explain’ SVM, KMC, DNN, RNN, 2AM, POMDP Computational Education Models Research Pierre Dillenbourg, EPFL

  13. K t B t Observable State Hidden State K t+1 B t+1 p(K t+1 | B t+1 , K t ) Bayesian Knowledge Tracing p(K t = ‘skill-x’ | B t = ‘correct answer’)= 1 - Guess p(K t = ‘skill-X’ | B t = ‘incorrect answer’)= 0 + Slip

  14. Improve the management of education systems Computational Education Models Research Pierre Dillenbourg, EPFL

  15. What about modeling learning outsider technology-based environments ?

  16. pre-requisites p(Succeed ( CS243 ) | Failed ( CS201 )) carreer p(Salary > T | { INF201, MA203,INF233,.. }) recommender 78% of those who select CS243 also selected CS411 ……. Campus Analytics Pierre Dillenbourg, EPFL

  17. how deep ?

  18. Relevant Behavioral Abstractions (Features) Education needs explainable AI Computational Education Models Research Pierre Dillenbourg, EPFL

  19. Relevant Behavioral Abstractions (Features) gaze(a)=ƒ(gaze(b))

  20. Gaze Recurrence

  21. Relevant Behavioral Abstractions gaze(listener)=ƒ(gaze(speaker)) Feature: Gaze recurrence Context: Collaborative learning

  22. Relevant Behavioral Abstractions gaze (learner) = ƒ (reference (teacher)) Feature: Withmeness Context: Lecturing

  23. Sarah d’Angelo, Kshitij Sharma, Darren Gergle, Pierre Dillenbourg (2016)

  24. Do finger-based or gaze-based deictics enhance learning ? Sarah d’Angelo, Kshitij Sharma, Darren Gergle, Pierre Dillenbourg (2016)

  25. Relevant Behavioral Abstractions gaze (learner) = ƒ (gaze (teacher)) Feature: ‘Withmeness’ Context: Lecturing

  26. K t B t Modeling in the wild ? Raca, Tormey & Dillenbourg

  27. Relevant Behavioral Abstractions gaze (learner) = ƒ (location (teacher)) Feature: Head rotations Context: Lecturing

  28. activity (teacher) = ƒ (gaze (teacher)) L. Prieto, K. Sharma, L. Kidzinsky, P. Dillenbourg

  29. Education brings nice challenges (1) Explainability Computational Education Models Research Pierre Dillenbourg, EPFL

  30. Education brings nice challenges (2) Cold Start Integrate expert’s knowledge Use simulation with synthetic students

  31. Education brings nice challenges (3) Exploration Exploitation Tradeoff Learner 1 A B Learner 2 Learner 3 A Learner 4 A Learner 5 Learner 6 A Learner 7 A Learner 8 ? Learner 9

  32. Education Context Exploration/ Explainable AI Cold Start Exploitation Trade-OFF Cohorte Simulations

  33. Multi-Armed Bandit (MAB) for Exploration-Exploitation ➢ Selecting learning activities ➢ LFA model ➢ Tested with simulated students ➢ Will present and discuss this work at ECTEL in Septembre 59 Louis Faucon, Pierre Dillenbourg, EPFL

  34. Education is a computational science EPFL Center for Learning Sciences

  35. GRAASP Happy Numbers SpeakUP

  36. Co Compu. u. Th Thinkin ing MO MOOCs Center La Lake of NC NCCR Piaget Pi Learning Co Collider Sciences EPFL Digital Education Ecosystem

  37. Social Sciences Humanities Education Modern Languages Psychology Old Language Sociology Littérature Ethnology Philosophy Anthropology Religion Economics Art Political Sciences Musicology Linguistics Museology History History This is not one science !!! Demography …. Management …..

  38. Education Docimology Didactics Instructional Psychology Instructional Design And…. Learning Technologies History of Education Sociology of Education Economy of Education Special Education Psychology Cognitive psychology Social psychology Psychometry Clinical psychology Differential Psychology Developmental Psychology Sociology

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend


More recommend