I am doctoral student at UC Berkeley in computer science (specialization: artificial intelligence), finishing in May 2014. I am honored to have been recently awarded a National Academy of Education/Spencer Dissertation Fellowship to support my final year of graduate school. My research focuses on applying probabilistic models to diagnosing student knowledge and choosing educational interventions. This has involved such projects as:

  • Diagnosing learners' beliefs from their actions: How can we interpret actions in complex environments like games and virtual labs, linking them to misunderstandings that a student may have? In this project, we use a variation of inverse reinforcement learning to infer how people's beliefs about the effects of their actions
  • Optimizing games for diagnosis: How can we automatically choose a design for a game so that players' actions will be most diagnostic about their beliefs? We search through a space of possible game designs to find versions that will give the most information about a person's beliefs. The resulting game might be used as an educational assessment or as a tool for cognitive science research; in either case, the optimized design will result in more efficient diagnosis.
  • Choosing pedagogical activities using POMDP planning: Which activities will be best for a learner, and how can we balance activities aimed at assessing the learner's knowledge with activities aimed at improving that knowledge? We have formulated teaching as a POMDP planning problem, and examined how the learner model and the model of the domain affect the optimal policy.
I enjoy collaborating with others on projects related to using computational techniques to improve educational technologies. I have been working with the WISE group on creating personalized feedback for students based on their performance on embedded assessments. My focus is on helping students better understand chemistry concepts, and as part of this project, I created an automatic scorer for student drawings of chemical reactions (a demo is available here). I am also involved with the ChemVLab+ project, where I work on data analysis and applying machine learning techniques to uncover what behaviors are associated with particular learning outcomes. Prior to my work on educational applications, I conducted experimental and theoretical work on language evolution and iterated learning at Berkeley.

I mainly work in the Computational Cognitive Science lab, and I'm advised by Tom Griffiths and Dan Klein.

I completed my undergraduate and masters' work at Stanford in Symbolic Systems, and following graduation, I worked in Stanford's Natural Language Processing Group as a research programmer for Professor Chris Manning. My master's thesis in Symbolic Systems involved creating computational models of students' learning in intelligent tutoring systems, and I was advised by Professor Dan Schwartz.

My other interests include feminist activism and improving access to science and math education for women and girls. In the past, I have volunteered with Girls Inc., teaching science to girls ages 6-14. I have also been involved in volunteering at Expanding Your Horizons events. I am originally from northern Minnesota and miss having a lake in my backyard - but the lack of snow in the Bay Area is a plus! Other interests include jigsaw puzzles, rollerblading and hiking, sci-fi books, and SET.

Resume [pdf]

Contact: rafferty AT cs DOT berkeley DOT edu


Ph.D. in progress in Computer Science at UC Berkeley, August 2008-present (Advanced to candidacy April 2012)

M.S., Computer Science, University of California, Berkeley, May 2011.

M.S., Symbolic Systems, Stanford University, June 2007.

B.S., Symbolic Systems, Focus in Artificial Intelligence, Stanford University, June 2007, with Distinction.

B.A., Feminist Studies, Focus in Women's Sexuality, Stanford University, June 2007, with Distinction.

Relevant Work Experience:

Research Programmer. Natural Language Processing Group, Computer Science, Stanford University, July 2007-August 2008.


Rafferty, Anna N., Thomas L. Griffiths, and Marc Ettlinger. (in press) “Greater learnability is not sufficient to produce cultural universals.” Cognition. [PDF]

Rafferty, Anna N., Jodi Davenport, and Emma Brunskill. (2013) “Estimating Student Knowledge from Paired Interaction Data.” Proceedings of The 6th International Conference on Educational Data Mining (EDM 2013). [PDF]

Rafferty, Anna N., Libby Gerard, Kevin McElhaney, Marcia C. Linn. (2013) “Automating Guidance for Students Chemistry Drawings.” Proceedings of Formative Feedback in Interactive Learning Environments (AIED Workshop). [PDF]

Rafferty, Anna N., Matei Zaharia, and Thomas L. Griffiths. (2012) “Optimally Designing Games for Cognitive Science Research.” Proceedings of The 34th Annual Conference of the Cognitive Science Society. p. 280-287. [PDF]

Rafferty, Anna N., Michelle L. LaMar, and Thomas L. Griffiths. (2012) “Inferring learners knowledge from observed actions.” Proceedings of The 5th International Conference on Educational Data Mining (EDM 2012). Winner of Best Poster Award. [PDF]

Davenport, Jodi, Anna Rafferty, Michael Timms, David Yaron, Michael Karabinos. (2012) “ChemVLab+: Evaluating a Virtual Lab Tutor for High School Chemistry.” Proceedings of The 10th International Conference of the Learning Sciences (ICLS 2012). [PDF]

Rafferty, Anna N., Emma Brunskill, Thomas L. Griffiths, and Patrick Shafto. (2011) “Faster teaching by POMDP planning.” Proceedings of The 15th International Conference on Artificial Intelligence in Education (AIED2011). p. 280-287. [PDF]

Rafferty, Anna N., Thomas L. Griffiths, and Marc Ettlinger. (2011) “Exploring the relationship between learnability and linguistic universals.” Proceedings of The 2nd Workshop on Cognitive Modeling and Computational Linguistics at ACL 2011. [PDF]

Rafferty, Anna N. and Thomas L. Griffiths. (2010) "Optimal language learning: The importance of starting representative." Proceedings of The 32nd Annual Conference of the Cognitive Science Society. [PDF]

Rafferty, Anna N., Thomas L. Griffiths, and Dan Klein. (2009) "Convergence Bounds for Language Evolution by Iterated Learning." Proceedings of The 31st Annual Conference of the Cognitive Science Society. [PDF]

Ramage, Daniel, Anna N. Rafferty, and Christopher D. Manning. (2009) "Random Walks for Text Semantic Similarity." Proceedings of ACL-IJCNLP TextGraphs-4 Workshop 2009. [PDF]

Rafferty, Anna N. and Christopher D. Manning. (2008) "Parsing Three German Treebanks: Lexicalized and Unlexicalized Baselines." Proceedings of Workshop on Parsing German, ACL-HLT 2008. [PDF]

de Marneffe, Marie-Catherine, Anna N. Rafferty, and Christopher D. Manning. (2008) "Finding Contradictions in Text." Proceedings of ACL-HLT 2008. [PDF]

Rafferty, Anna N. and Michael Yudelson. (20007) "Applying Learning Factors Analysis to Build Stereotypic Student Models." Proceedings of Artificial Intelligence in Education, 2007. Winner of Best Paper Award for the Young Researcher Track. [PDF]

Marie-Catherine de Marneffe, Bill MacCartney, Trond Grenager, Daniel Cer, Anna Rafferty and Christopher D. Manning. (2006) "Learning to distinguish valid textual entailments." Proceedings of The Second PASCAL Challenges Workshop. 2006. [PDF]

Master's Thesis:

"Using FACT to Challenge Assumptions: Frequency, Accuracy, Choice, and Timing in Machine Learning." Symbolic Systems Department, Stanford University, June 2007. Advised by Professor Dan Schwartz, with second reader Professor Ken Koedinger. [PDF]