Publications
(Email me at ebrun AT cs dot cmu dot edu for any papers listed without links)
2017
- Coming soon: 2 AAAI papers, 1 IAAI paper, 1 Learning at Scale paper
2016
- Data-Efficient Off-Policy Policy Evaluation for Reinforcement Learning
[link,
reviews]
P.Thomas and E.Brunskill
to appear in International Conference on Machine Learning (ICML) 2016
- Energetic Natural Gradient Descent
[link,
reviews]
P.Thomas, C.Dann, B. Castro da Silva and E.Brunskill
to appear in International Conference on Machine Learning (ICML) 2016
- Efficient Bayesian Clustering for Reinforcement Learning
[link]
Travis Mandel, Yun-En Liu, Emma Brunskill and Zoran Popovic
to appear in International Joint Conference on AI (IJCAI) 2016
- Latent Contextual Bandits and their Application to Personalized Recommendations for New Users
[pdf]
Li Zhou and Emma Brunskill
to appear in International Joint Conference on AI (IJCAI) 2016
- PAC Continuous State Online Multitask Reinforcement Learning with Identification
[link,
supplemental]
Yao Liu, Zhaohan Guo, and Emma Brunskill
AAMAS 2016
- A PAC RL Algorithm for Episodic POMDPs
[link]
Zhaohan Guo, Shayan Doroudi and Emma Brunskill
AISTATS 2016
- Offline Evaluation of Online Reinforcement Learning Algorithms [link]
Travis Mandel, Yun-En Liu, Emma Brunskill and Zoran Popovic
AAAI 2016
- Sequence Matters, But How Exactly? A Methodology for Evaluating Activity Sequences from Data
Shayan Doroudi, Kenneth Holstein, Vincent Aleven and Emma Brunskill
to appear in Educational Data Mining (EDM) 2016
- Questimator: Generating Knowledge Assessments for Arbitrary Topics
Qi Guo, Chinmay Kulkarni, Aniket Kittur, Jeffrey Bigham and Emma Brunskill
to appear in International Joint Conference on AI (IJCAI) 2016
- Interface Design Optimization as a Multi-Armed Bandit Problem
James Lomas, Jodi Forlizzi, Nikhil Poonwala, Nirmal Patel, Sharan Shodhan, Kishan Patel, Ken Koedinger, and Emma Brunskill
to appear in Computer Human Interaction (CHI) 2016.
- Towards a Learning Science for Complex Crowdsourcing Tasks
Shayan Doroudi, Ece Kamar, Emma Brunskill and Eric Horvitz
to appear in Computer Human Interaction (CHI) 2016.
- Automatically Learning to Teach to the Learning Objective
[pdf]
Rika Antonova, Joe Runde, Dexter Lee, and Emma Brunskill
work in progress to appear in Learning at Scale 2016
- Faster Teaching via POMDP Planning
Anna Rafferty, Emma Brunskill, Tom Griffiths and Pat Shafto
accepted Cognitive Science
2015
- Sample Complexity of Episodic Fixed-Horizon Reinforcement Learning
[arxiv pdf]
Christoph Dann and Emma Brunskill
NIPS 2015
- Concurrent PAC RL
[pdf]
Zhaohan (Daniel) Guo and Emma Brunskill.
AAAI 2015
- The Queue Method: Handling Delay, Heuristics, Prior Data, and Evaluation in
Bandits[link]
Travis Mandel, Yun-En Liu, Emma Brunskill, Zoran Popovic.
AAAI 2015
- From Predictive Models to Instructional Policies
[pdf soon]
Joseph Rollinson and Emma Brunskill
Educational Data Mining (EDM) 2015.
- Towards Understanding How to Leverage Sense-making, Induction/Refinement and Fluency to Improve Robust Learning
[pdf soon]
Shayan Doroudi, Kenneth Holstein, Vincent Aleven and Emma Brunskill.
Educational Data Mining (EDM) 2015.
- Learning the Features Used to Decide How to Teach
[link]
Min Hyung Lee, Joe Runde, Warfa Jabril, Zhouying Wang, and Emma Brunskill.
Learning at Scale 2015. Work in Progess paper.
2014
- PAC-inspired Option Discovery in Lifelong Reinforcement Learning
[pdf]
Note:Lazaric and Fruit pointed out that PAC-SMDP and PAC-MDP are not always directly comparable, hence our paper's analysis of the benefit of options over primitive actions does not always hold.
E.Brunskill and L.Li
ICML 2014
- Resource-Efficient Stochastic Optimization of a Locally Smooth Function under Correlated Bandit Feedback
[pdf]
M.Azar, A.Lazaric and E.Brunskill
ICML 2014
- Trading Off Scientific Knowledge and User Learning with Multi-Armed Bandits
[pdf]
Yun-En Liu, Travis Mandel, Emma Brunskill, and Zoran Popovic.
Educational Data Mining (EDM) 2014.
- Towards Automatic Experimentation of Educational Knowledge
Nominated for best paper [pdf]
Yun-En Liu, Travis Mandel, Emma Brunskill, and Zoran Popovic.
Computer Human Interaction (CHI) 2014.
- Offline Policy Evaluation Across Representations with Applications to Educational Games [pdf]
Travis Mandel, Yun-En Liu, Sergey Levine, Emma Brunskill, Zoran Popovic.
Autonomous Agents and Multi-Agent Systems (AAMAS) 2014.
2013
- Sequential Transfer in Multi-armed Bandit with Finite Set of Models [link]
M.Azar, A.Lazaric and E.Brunskill
NIPS 2013
- Regret Bounds for Reinforcement Learning with Policy Advice [link]
M.Azar, A.Lazaric and E. Brunskill
European Conference on Machine Learning (ECML) 2013
- Sample Complexity of Multi-task Reinforcement Learning. [pdf]
E.Brunskill and L.Li
Conference on Uncertainty in Artificial Intelligence (UAI) 2013
- Understanding Sequential Decisions via Inverse Reinforcement Learning.
Siyuan Liu, Miguel Araujo, Emma Brunskill, Rosaldo Rossetti, Joao Barros, Ramayya Krishnan
MDM 2013
- Predicting Player Moves in an Educational Game: A Hybrid Approach*
Nominated for best paper [pdf]
Yun-En Liu, Travis Mandel, Eric Butler, Erik Andersen, Eleanor O'Rourke, Emma Brunskill, Zoran Popovic.
Educational Data Mining (EDM) 2013.
- Estimating Student Knowledge from Paired Interaction Data. [pdf]
A.Rafferty, J.Davenport, and E.Brunskill
Educational Data Mining (EDM) 2013
- New Potentials for Data-driven Intelligent Tutoring System Development and Optimization.
Koedinger, K.R., Brunskill, E., Baker, R.S.J.d., McLaughlin, E.A., Stamper, J.
AI Magazine
- Towards Operationalizing Outlier Detection in Community Health Programs
T.McCarthy, B.DeRenzi, J.Blumenstock and E.Brunskill
to appear as Note in Information Communications Technology for International Development (ICTD) 2013
- Analysis of the Impact of Errors Made During Health Data Collection Using Mobile Phones: Exploring Error Modeling and Automatic Diagnosis.
S.Palkar and E.Brunskill
ACM DEV 2013 (Poster)
2012
- Incentive Decision Processes
[pdf]
S.Reddi and E.Brunskill
Conference on Uncertainty in Artificial Intelligence (UAI) 2012
- The Impact on Individualizing Student Models on Necessary Practice
Opportunities
*Nominated for best paper
[pdf]
J.I.Lee and E.Brunskill
International Conference on Educational Data Mining (EDM) 2012
- Policy Building -- An Extension To User Modeling
[pdf]
M.Yudelson and E.Brunskill
International Conference on Educational Data Mining (EDM) 2012
- Bayes-optimal reinforcement learning for discrete uncertainty domains.
E.Brunskill
AAMAS 2012 (Extended Abstract)
- Global and regional hearing impairment prevalence: an analysis of 42 studies in 29 countries [link]
G.Stevens, S.Flaxman, E.Brunskill, M. Mascarenhas, C.Mathers, M.Finucane
The European Journal of Public Health
2011 and prior
- Estimating prerequisite structure from noisy data
[pdf]
E.Brunskill
International Conference on Educational Data Mining (EDM) 2011
- Partially observable sequential decision making for problem selection in an intelligent tutoring system
[pdf]
E.Brunskill and S.Russell
poster in International Conference on Educational Data Mining (EDM) 2011
- Faster teaching by POMDP planning
[pdf]
A.Rafferty, E.Brunskill, T.Griffiths, and P.Shafto
International Conference on Artificial Intelligence in Education (AIED) 2011
- Designing mobile interfaces for novice and low-literacy users
[pdf]
I. Medhi, S. Patnaik, E.Brunskill, S. N. Nagasena Gautama, W. Thies, and K. Toyama
ACM Transactions on Computer-Human Interaction 2011
- Efficient planning under uncertainty with macro-actions
[pdf]
R.He, E. Brunskill and N. Roy
Journal of Artificial Intelligence Research 2011
- Evaluating an adaptive multi-user educational tool for low-resource regions
[pdf]
E.Brunskill, S.Garg, C.Tseng, J.Pal and L.Findlater
International Conference on Information and Communication Technologies and Development (ICTD) 2010
- RAPID: A reachable anytime planner for imprecisely-sensed domains
[pdf]
E.Brunskill and S.Russell
Uncertainty in Artificial Intelligence (UAI) 2010
- PUMA: planning under uncertainty with macro-actions
[pdf]
R.He, E.Brunskill and N.Roy
AAAI Conference on Artificial Intelligence (AAAI) 2010
- Planning in partially-observable switching-mode continuous domains
[pdf]
E.Brunskill, L.Kaelbling, T.Lozano-Perez, and N.Roy
Annals of Mathematics and Artificial Intelligence 2010
- When policies can be trusted: analyzing a criteria to identify optimal policies in MDPs with unknown model parameters
[pdf]
(This corrected version makes a 2 sentence clarification that only the final DeltaQ represents an estimate of g)
E.Brunskill
International Conference on Automated Planning and Scheduling (ICAPS) 2010
- Provably efficient learning with typed parametric models
[pdf]
E.Brunskill, B. Leffler, L. Li, M. Littman, and N. Roy.
Journal of Machine Learning Research 2009
- Evaluating the accuracy of data collection on mobile phones: a study of forms, SMS, and voice
[pdf]
S.Patnaik, E.Brunskill, and W.Thies.
International Conference on Information and Communication Technologies and Development (ICTD) 2009
- Where to go: interpreting natural directions using global inference
[pdf]
Y.Wei, E.Brunskill, T.Kollar and N.Roy.
International Conference on Robotics and Automation (ICRA) 2009
- CORL: a continuous-state offset-dynamics reinforcement learner
[pdf]
E.Brunskill, B.Leffler, L.Li, M.Littman and N.Roy.
Uncertainty in Artificial Intelligence (UAI) 2008
-
Perceptual switch rates with ambiguous structure-from-motion figures in bipolar disorder
[pdf].
K.Krug, E.Brunskill, A.Scarna, G.Goodwin and A.Parker.
Proceedings of the Royal Society B 2008.
- A supervised learning approach for collision detection in legged locomotion.
[pdf]
F.Doshi, E.Brunskill, A.Shkolnik, T.Kollar, K.Rohanimanesh, R.Tedrake, and N.Roy
International Conference on Intelligent Robots and Systems (IROS) 2007
- Topological mapping using spectral clustering and classification
[pdf]
E.Brunskill, T.Kollar and N.Roy
International Conference on Intelligent Robots and Systems (IROS) 2007
- Adaptive state space construction with reinforcement learning for robots.
E.Brunskill, E.Uchibe, and K.Doya
poster International Conference on Robotics and Automation (ICRA) 2006
- SLAM using incremental probabilistic PCA and dimensionality reduction.
[pdf]
E.Brunskill and N.Roy.
International Conference on Robotics and Automation (ICRA) 2005
Workshop and Symposium Papers
- Routing for rural health: optimizing community health worker visit schedules
[pdf]
E.Brunskill and N.Lesh
position paper AAAI Spring Symposium on Artificial Intelligence for Development 2010
- Learning to identify locally actionable health anomalies
[pdf]
K.Chen, E.Brunskill, J.Dick and P.Dhadialla
position paper AAAI Spring Symposium on Artificial Intelligence for Development 2010
- How close is close enough? Finding optimal policies in PAC-style reinforcement learning.
E. Brunskill
Abstract
NIPS 2008 Workshop on Model Uncertainty and Risk in Reinforcement Learning
- Continuous-state POMDPs with hybrid dynamics.
[pdf]
E.Brunskill, L.Kaelbling, T.Lozano-Perez, and N.Roy.
Proceedings of the International Symposium on Artificial Intelligence and Mathematics (ISAIM) 2008
- Continuous state POMDPs for object manipulation tasks.
E.Brunskill
Association for the Advancement of Artificial Intelligence (AAAI) 2007 Doctoral Consortium
- Lessons from prototyping a microfinance distance learning tool.
[pdf]
E.Brunskill and T.Parikh
CHI workshop on User Centered Design for International Development 2007
- Building peer-to-peer systems with Chord, a distributed lookup service.
F.Dabek, E.Brunskill, F.Kaashoek, D.Karger, R.Morris, I.Stoica, and H.Balakrishnan.
in Proceedings of the 8th Workshop on Hot Topics in Operating Systems (HotOS-VIII) 2001
Ph.D. Thesis
Other Publications
- LittleDog learning locomotion project
CSAIL research abstract 2006
- Impact of Bipolar Disorder on Ambiguous Structure-from-Motion Percepts
E.Brunskill, et al.
Society for Neuroscience Abstract (SFN) 2003