NIPS 2012 Workshop
Personalizing Education With Machine Learning

Saturday December 8, 2012
Harvey's Convention Center, Emerald Bay 1
Lake Tahoe, California


Michael Mozer
Department of Computer Science & Institute of Cognitive Science
University of Colorado Boulder

Javier Movellan
Institute for Neural Computation
University of California San Diego

Robert Lindsey
Department of Computer Science
University of Colorado Boulder

Jacob Whitehill
Department of Computer Science and Engineering
University of California San Diego

Workshop Description

The field of education has the potential to be transformed by the internet and intelligent computer systems. Evidence for the first stage of this transformation is abundant, from the Stanford online AI and Machine Learning courses to web sites such as Kahn Academy that offer on line lessons and drills.  However, the delivery of instruction via web-connected devices is merely a precondition for what may become an even more fundamental transformation:  the personalization of education.

In traditional classroom settings, teachers must divide their attention and time among many students and hence have limited ability to observe and customize instruction to individuals. Even in one-on-one tutoring sessions, teachers rely on intuition and experience to choose the material and stye of instruction that they believe would provide the greatest benefit given the student's current state of understanding.  

In order both to assist human teachers in traditional classroom environments and to improve automated tutoring systems to match the capabilities of expert human tutors, one would like to develop formal approaches that can:

Machine learning provides a rich set of tools, extending classical psychometric approaches, for data-driven latent state inference, policy optimization, and personalization.  Years ago, it would have been difficult to obtain enough data for a machine learning approach. However, online interactions with students have become commonplace, and these interactions yield a wealth of data. The data to be mined go beyond what is typed: Cameras and microphones are ubiquitous on portable devices, allowing for the exploitation of subtle video and audio cues.  Because web-based instruction offers data from a potentially vast collection of diverse learners, the population of learners should serve useful in drawing inferences about individual learners.

Mining the vast datasets on teaching and learning that emerge over the coming years may both yield important insights into effective teaching strategies and also deliver practical tools to assist both human and automated teachers.

The goal of this workshop is to bring together researchers in machine learning, data mining, and computational statistics with researchers in education, psychometrics, intelligent tutoring systems, and designers of web-based instructional software.  Although a relatively young journal and conference on educational data mining has been established (, there are exciting opportunities to enhance interactions between cognitive scientists interested in education and machine learning theoreticians.

Potential Participants

We hope to draw participants from diverse academic backgrounds, including:
  • machine learning theoreticians interested in formal approaches to teaching from a computational perspective
  • AI researchers interested in computer vision and EEG to recover information about an individual's affective and mental state
  • established researchers in intelligent tutoring systems
  • psychologists studying practical aspects of human learning and memory 
  • developers of web sites that collect large volumes of student data
  • distinguished educators, including educators who can discuss the current state of the classroom, and educators in the vanguard of the e-ducation revolution 

Schedule Overview

Topic Time Presenter Title And Abstract Slides
Introduction 7:30-7:45 Michael Mozer (University of Colorado)
Javier Movellan (, UCSD)
Doris Alvarez (Director, TDLC Educator Network; Founding Principal, Preuss School)
Harold Pashler (UCSD)
Policy optimization 7:45-8:05 Jerry Zhu (Wisconsin) A Computational Teaching Theory For Bayesian Learners
8:10-8:30 Emma Brunskill (CMU) Pedagogical Activity Selection: Drawing Insight From Sequential Decision Making Under Uncertainty
8:35-8:55 Jacob Whitehill (, UCSD) A Stochastic Optimal Control Perspective on Affect-Sensitive Teaching pdf
9:00-9:30 poster session and coffee break
9:30-9:45 Min Chi  (Stanford) Empirically Evaluating the Application of Reinforcement Learning to the Induction of Effective and Adaptive Pedagogical Strategies
Video analysis 9:45-10:00 Vikram Ramanarayanan (USC) A Framework For Automated Analysis Of Videos Of Informal Classroom Educational Settings
MOOCs 10:00-10:20 Andrew Ng  (Stanford, The Online Revolution: Education For Everyone ppt
    10:30-11:00 general discussion; continuation of poster session
José  González-Brenes (CMU) Topical HMMs for Factorization of Input-Output Sequential Data
William Jenkins (Scientific Learning) What Young Children Need To Learn About Numbers: Differences In Learning Style And Response To Error Correction In Pre-Kindergarten And Kindergarten Students pdf
Eun-Sol Kim (Seoul National University) Learning-Style Recognition From Eye-Hand Movement Using A Dynamic Bayesian Network
Andrew Lan (Rice) Joint Sparse Factor Analysis And Topic Modeling For Learning Analytics
Nan Li (CMU) Automated Creation Of Intelligent Tutoring To Support Personalized Online Learning
Loizos Michael (Open University of Cyprus) “Primum Non Nocere” For Personalized Education
Arti Ramesh (Maryland) User Role Prediction In Online Discussion Forums Using Probabilistic Soft Logic pdf
Latent state inference 15:30-15:50 Richard Scheines (CMU) Machine Learning, Causal Model Search, and Educational Data
15:55-16:10 April Galyardt (University of Georgia) Modeling Student Strategy Usage With Mixed Membership Models pdf
Anna Rafferty (UC Berkeley)
Using Inverse Reinforcement Learning To Diagnose Learners' Misconceptions
16:40-16:55 Yanbo Xu (CMU) A Dynamic Higher-Order DINA Model To Trace Multiple Skills
17:00-17:20 poster session and coffee break
17:20-17:35 Vivienne Ming (Socos LLC, Berkeley) and Norma Ming (Nexus, Berkeley) Inferring Conceptual Knowledge From Unstructured Student Writing ppt
17:40-18:00 Andrew Waters (Rice) Learning Analytics Via Sparse Factor Analysis ppt
18:05-18:25 Robert Lindsey (University of Colorado) Inferring History-Dependent Memory Strength Via Collaborative Filtering: Toward The Optimization Of Long-Term Retention pdf
18:30-19:00 general discussion


The workshop is sponsored in part by the UCSD Temporal Dynamics of Learning Center, which is funded by the National Science Foundation.   TDLC Logo