TY - GEN
T1 - Faster teaching by POMDP planning
AU - Rafferty, Anna N.
AU - Brunskill, Emma
AU - Griffiths, Thomas L.
AU - Shafto, Patrick
N1 - Copyright:
Copyright 2011 Elsevier B.V., All rights reserved.
PY - 2011
Y1 - 2011
N2 - Both human and automated tutors must infer what a student knows and plan future actions to maximize learning. Though substantial research has been done on tracking and modeling student learning, there has been significantly less attention on planning teaching actions and how the assumed student model impacts the resulting plans. We frame the problem of optimally selecting teaching actions using a decision-theoretic approach and show how to formulate teaching as a partially-observable Markov decision process (POMDP) planning problem. We consider three models of student learning and present approximate methods for finding optimal teaching actions given the large state and action spaces that arise in teaching. An experimental evaluation of the resulting policies on a simple concept-learning task shows that framing teacher action planning as a POMDP can accelerate learning relative to baseline performance.
AB - Both human and automated tutors must infer what a student knows and plan future actions to maximize learning. Though substantial research has been done on tracking and modeling student learning, there has been significantly less attention on planning teaching actions and how the assumed student model impacts the resulting plans. We frame the problem of optimally selecting teaching actions using a decision-theoretic approach and show how to formulate teaching as a partially-observable Markov decision process (POMDP) planning problem. We consider three models of student learning and present approximate methods for finding optimal teaching actions given the large state and action spaces that arise in teaching. An experimental evaluation of the resulting policies on a simple concept-learning task shows that framing teacher action planning as a POMDP can accelerate learning relative to baseline performance.
UR - http://www.scopus.com/inward/record.url?scp=79959322300&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=79959322300&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-21869-9_37
DO - 10.1007/978-3-642-21869-9_37
M3 - Conference contribution
AN - SCOPUS:79959322300
SN - 9783642218682
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 280
EP - 287
BT - Artificial Intelligence in Education - 15th International Conference, AIED 2011
T2 - 15th International Conference on Artificial Intelligence in Education, AIED 2011
Y2 - 28 June 2011 through 1 July 2011
ER -