Emma Brunskill
Associate Professor of Computer Science and, by courtesy, of Education
Academic Appointments
-
Associate Professor, Computer Science
-
Associate Professor (By courtesy), Graduate School of Education
-
Faculty Affiliate, Institute for Human-Centered Artificial Intelligence (HAI)
Honors & Awards
-
Alumni Impact Award, University of Washington Computer Science (2020)
-
Young Investigator Award, Office of Naval Research (2015)
-
CAREER Award, NSF (2014)
-
Faculty Fellowship, Microsoft (2012)
Program Affiliations
-
Symbolic Systems Program
Professional Education
-
PhD, Massachusetts Institute of Technology, Computer Science (2009)
2024-25 Courses
- Counterfactuals: The Science of What Ifs?
CS 31N (Spr) - Lean Launchpad for Education
EDUC 260 (Aut) - Reinforcement Learning
CS 234 (Win) -
Independent Studies (11)
- Advanced Reading and Research
CS 499 (Aut, Win, Spr, Sum) - Advanced Reading and Research
CS 499P (Aut, Win, Spr, Sum) - Curricular Practical Training
CS 390A (Aut, Win, Spr, Sum) - Curricular Practical Training
CS 390B (Aut, Win, Spr, Sum) - Independent Project
CS 399 (Aut, Win, Spr, Sum) - Independent Project
CS 399P (Aut, Win, Spr, Sum) - Independent Work
CS 199 (Aut, Win, Spr, Sum) - Independent Work
CS 199P (Aut, Win, Spr, Sum) - Part-time Curricular Practical Training
CS 390D (Aut, Win, Spr, Sum) - Supervised Undergraduate Research
CS 195 (Aut, Win, Spr, Sum) - Writing Intensive Senior Research Project
CS 191W (Aut, Win, Spr)
- Advanced Reading and Research
-
Prior Year Courses
2023-24 Courses
- Reinforcement Learning
CS 234 (Spr)
2022-23 Courses
- Advanced Survey of Reinforcement Learning
CS 332 (Aut) - Counterfactuals: The Science of What Ifs?
CS 31N (Spr) - Reinforcement Learning
CS 234 (Win)
2021-22 Courses
- Causality, Counterfactuals and AI
OSPOXFRD 48 (Spr) - Reinforcement Learning
CS 234 (Win)
- Reinforcement Learning
Stanford Advisees
-
Doctoral Dissertation Reader (AC)
Lauren Gillespie, Saurabh Kumar, Gabriel Poesia Reis e Silva, Roshni Sahoo, Garrett Thomas, Rose Wang -
Postdoctoral Faculty Sponsor
Ge Gao, Zhaoqi Li -
Master's Program Advisor
Jenny Chen, Mehmet Hamza Erol, Audrey Kwan, Peyton Lee, JB Jong Beom Lim, Alex Paek, Arpit Ranasaria, Megan Santhumayor, Akhil Vyas, Nick Walker, Evelyn Yee -
Doctoral Dissertation Co-Advisor (AC)
Aishwarya Mandyam, Allen Nie, Henry Zhu -
Doctoral (Program)
Joy He-Yueya, Alex Nam
All Publications
-
Reinforcement learning tutor better supported lower performers in a math task
MACHINE LEARNING
2024
View details for DOI 10.1007/s10994-023-06423-9
View details for Web of Science ID 001159435300001
-
Texting and tutoring: Short-term K-3 reading interventions during the pandemic
JOURNAL OF EDUCATIONAL RESEARCH
2023
View details for DOI 10.1080/00220671.2023.2251432
View details for Web of Science ID 001059545700001
-
Constraint Sampling Reinforcement Learning: Incorporating Expertise For Faster Learning
ASSOC ADVANCEMENT ARTIFICIAL INTELLIGENCE. 2022: 7841-7849
View details for Web of Science ID 000893639100088
-
Power Constrained Bandits.
Proceedings of machine learning research
1800; 149: 209-259
Abstract
Contextual bandits often provide simple and effective personalization in decision making problems, making them popular tools to deliver personalized interventions in mobile health as well as other health applications. However, when bandits are deployed in the context of a scientific study-e.g. a clinical trial to test if a mobile health intervention is effective-the aim is not only to personalize for an individual, but also to determine, with sufficient statistical power, whether or not the system's intervention is effective. It is essential to assess the effectiveness of the intervention before broader deployment for better resource allocation. The two objectives are often deployed under different model assumptions, making it hard to determine how achieving the personalization and statistical power affect each other. In this work, we develop general meta-algorithms to modify existing algorithms such that sufficient power is guaranteed while still improving each user's well-being. We also demonstrate that our meta-algorithms are robust to various model mis-specifications possibly appearing in statistical studies, thus providing a valuable tool to study designers.
View details for PubMedID 34927078
-
EnglishRot: An Al-Powered Conversational System for Second Language Learning
ASSOC COMPUTING MACHINERY. 2021: 434-444
View details for DOI 10.1145/3397481.3450648
View details for Web of Science ID 000747690200052
-
Automatic Adaptive Sequencing in a Webgame
SPRINGER INTERNATIONAL PUBLISHING AG. 2021: 430-438
View details for DOI 10.1007/978-3-030-80421-3_47
View details for Web of Science ID 000718916000047
-
Learning When-to-Treat Policies
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
2020
View details for DOI 10.1080/01621459.2020.1831925
View details for Web of Science ID 000596368100001
-
Scaling up behavioral science interventions in online education.
Proceedings of the National Academy of Sciences of the United States of America
2020
Abstract
Online education is rapidly expanding in response to rising demand for higher and continuing education, but many online students struggle to achieve their educational goals. Several behavioral science interventions have shown promise in raising student persistence and completion rates in a handful of courses, but evidence of their effectiveness across diverse educational contexts is limited. In this study, we test a set of established interventions over 2.5 y, with one-quarter million students, from nearly every country, across 247 online courses offered by Harvard, the Massachusetts Institute of Technology, and Stanford. We hypothesized that the interventions would produce medium-to-large effects as in prior studies, but this is not supported by our results. Instead, using an iterative scientific process of cyclically preregistering new hypotheses in between waves of data collection, we identified individual, contextual, and temporal conditions under which the interventions benefit students. Self-regulation interventions raised student engagement in the first few weeks but not final completion rates. Value-relevance interventions raised completion rates in developing countries to close the global achievement gap, but only in courses with a global gap. We found minimal evidence that state-of-the-art machine learning methods can forecast the occurrence of a global gap or learn effective individualized intervention policies. Scaling behavioral science interventions across various online learning contexts can reduce their average effectiveness by an order-of-magnitude. However, iterative scientific investigations can uncover what works where for whom.
View details for DOI 10.1073/pnas.1921417117
View details for PubMedID 32541050
-
Sublinear Optimal Policy Value Estimation in Contextual Bandits
ADDISON-WESLEY PUBL CO. 2020: 4377–86
View details for Web of Science ID 000559931301082
-
Being Optimistic to Be Conservative: Quickly Learning a CVaR Policy
ASSOC ADVANCEMENT ARTIFICIAL INTELLIGENCE. 2020: 4436-4443
View details for Web of Science ID 000667722804062
-
Fake It Till You Make It: Learning-Compatible Performance Support
JMLR-JOURNAL MACHINE LEARNING RESEARCH. 2020: 915-924
View details for Web of Science ID 000722423500084
-
Off-Policy Policy Gradient with State Distribution Correction
JMLR-JOURNAL MACHINE LEARNING RESEARCH. 2020: 1180-1190
View details for Web of Science ID 000722423500109
-
Interpretable Off-Policy Evaluation in Reinforcement Learning by Highlighting Influential Transitions
JMLR-JOURNAL MACHINE LEARNING RESEARCH. 2020
View details for Web of Science ID 000683178503072
-
Supporting Children's Math Learning with Feedback-Augmented Narrative Technology
ASSOC COMPUTING MACHINERY. 2020: 567-580
View details for DOI 10.1145/3392063.3394400
View details for Web of Science ID 000675620600050
-
Towards the Systematic Reporting of the Energy and Carbon Footprints of Machine Learning
JOURNAL OF MACHINE LEARNING RESEARCH
2020; 21
View details for Web of Science ID 000608918500001
-
Frequentist Regret Bounds for Randomized Least-Squares Value Iteration
ADDISON-WESLEY PUBL CO. 2020: 1954–63
View details for Web of Science ID 000559931304004
-
Where's the Reward?: A Review of Reinforcement Learning for Instructional Sequencing
INTERNATIONAL JOURNAL OF ARTIFICIAL INTELLIGENCE IN EDUCATION
2019; 29 (4): 568–620
View details for DOI 10.1007/s40593-019-00187-x
View details for Web of Science ID 000504748200005
-
Preventing undesirable behavior of intelligent machines.
Science (New York, N.Y.)
2019; 366 (6468): 999–1004
Abstract
Intelligent machines using machine learning algorithms are ubiquitous, ranging from simple data analysis and pattern recognition tools to complex systems that achieve superhuman performance on various tasks. Ensuring that they do not exhibit undesirable behavior-that they do not, for example, cause harm to humans-is therefore a pressing problem. We propose a general and flexible framework for designing machine learning algorithms. This framework simplifies the problem of specifying and regulating undesirable behavior. To show the viability of this framework, we used it to create machine learning algorithms that precluded the dangerous behavior caused by standard machine learning algorithms in our experiments. Our framework for designing machine learning algorithms simplifies the safe and responsible application of machine learning.
View details for DOI 10.1126/science.aag3311
View details for PubMedID 31754000
-
Fairer but Not Fair Enough On the Equitability of Knowledge Tracing
ASSOC COMPUTING MACHINERY. 2019: 335–39
View details for DOI 10.1145/3303772.3303838
View details for Web of Science ID 000473277300044
-
PLOTS: Procedure Learning from Observations using Subtask Structure
ASSOC COMPUTING MACHINERY. 2019: 1007–15
View details for Web of Science ID 000474345000116
-
Almost Horizon-Free Structure-Aware Best Policy Identification with a Generative Model
NEURAL INFORMATION PROCESSING SYSTEMS (NIPS). 2019
View details for Web of Science ID 000534424305060
-
Limiting Extrapolation in Linear Approximate Value Iteration
NEURAL INFORMATION PROCESSING SYSTEMS (NIPS). 2019
View details for Web of Science ID 000534424305059
-
Offline Contextual Bandits with High Probability Fairness Guarantees
NEURAL INFORMATION PROCESSING SYSTEMS (NIPS). 2019
View details for Web of Science ID 000535866906056
-
Value Driven Representation for Human-in-the-Loop Reinforcement Learning
ASSOC COMPUTING MACHINERY. 2019: 176–80
View details for DOI 10.1145/3320435.3320471
View details for Web of Science ID 000482185300025
-
QuizBot: A Dialogue-based Adaptive Learning System for Factual Knowledge
ASSOC COMPUTING MACHINERY. 2019
View details for DOI 10.1145/3290605.3300587
View details for Web of Science ID 000474467904049
-
Key Phrase Extraction for Generating Educational Question-Answer Pairs
ASSOC COMPUTING MACHINERY. 2019
View details for DOI 10.1145/3330430.3333636
View details for Web of Science ID 000507611000020
-
BookBuddy: Turning Digital Materials Into Interactive Foreign Language Lessons Through a Voice Chatbot
ASSOC COMPUTING MACHINERY. 2019
View details for DOI 10.1145/3330430.3333643
View details for Web of Science ID 000507611000030
-
Shared Autonomy for an Interactive AI System
ASSOC COMPUTING MACHINERY. 2018: 20–22
View details for DOI 10.1145/3266037.3266088
View details for Web of Science ID 000494261200007
-
Representation Balancing MDPs for Off-Policy Policy Evaluation
NEURAL INFORMATION PROCESSING SYSTEMS (NIPS). 2018
View details for Web of Science ID 000461823302064
-
Unifying PAC and Regret: Uniform PAC Bounds for Episodic Reinforcement Learning
NEURAL INFORMATION PROCESSING SYSTEMS (NIPS). 2017
View details for Web of Science ID 000452649405077
-
Regret Minimization in MDPs with Options without Prior Knowledge
NEURAL INFORMATION PROCESSING SYSTEMS (NIPS). 2017
View details for Web of Science ID 000452649403023
-
Using Options and Covariance Testing for Long Horizon Off-Policy Policy Evaluation
NEURAL INFORMATION PROCESSING SYSTEMS (NIPS). 2017
View details for Web of Science ID 000452649402053