Emma Brunskill

Emma Brunskill is an American computer scientist and professor specializing in reinforcement learning, machine learning, and sequential decision-making. She is known for research on offline policy evaluation, contextual bandits, and safe exploration, contributing to applications in healthcare, education (discipline), and robotics. Brunskill's work intersects with institutions such as Stanford University, the Massachusetts Institute of Technology, and conferences like NeurIPS and ICML.

Early life and education

Brunskill completed undergraduate studies at a major institution before pursuing graduate research at Stanford University and doctoral work at the Massachusetts Institute of Technology. Her doctoral advisors and collaborators have included faculty affiliated with MIT Computer Science and Artificial Intelligence Laboratory and Stanford Artificial Intelligence Laboratory. During her formative years she engaged with research groups that have ties to projects connected to DARPA, Netflix Prize, and industry labs like Google Research and Microsoft Research.

Research and career

Brunskill's career spans academic appointments, research fellowships, and collaborations with technology companies and government-funded programs. She has held positions at universities and research centers that collaborate with organizations such as National Science Foundation, Defense Advanced Research Projects Agency, and interdisciplinary initiatives involving Harvard University and University of California, Berkeley. Her research agenda emphasizes practical algorithms for offline policy evaluation, exploration strategies in reinforcement learning, and scalable methods for contextual bandits. Brunskill regularly publishes at venues including NeurIPS, ICML, AAAI, UAI, and AAMAS, and participates in workshops tied to IJCAI and the Association for Computing Machinery.

Major contributions and publications

Brunskill has produced influential papers on topics such as offline policy evaluation, sample-efficient reinforcement learning, and robust estimation for sequential decision processes. Her work addresses challenges in applying reinforcement learning to domains like healthcare decision-making, adaptive tutoring systems related to Carnegie Mellon University research, and human-in-the-loop robotics linked to labs at MIT and Stanford University. Notable contributions include algorithms and theoretical analyses that improve confidence bounds for policy value estimation, methods for efficient exploration under safety constraints, and approaches for contextual bandits with rich observational data. Her publications have been cited in the context of follow-on work from researchers at University of Washington, University of Toronto, University College London, and industry teams at Amazon, Facebook AI Research, and DeepMind.

Awards and honors

Brunskill's research has been recognized by awards and fellowships from organizations such as the National Science Foundation, leading foundations tied to computer science research, and conference best-paper recognitions at venues such as NeurIPS and ICML. She has received research grants and early-career awards comparable to those given by institutions like ACM and the IEEE, and has been invited to speak at symposia organized by AAAI, NIPS retrospectives, and workshops hosted by Google Research and Microsoft Research.

Teaching and mentorship

As a professor and mentor, Brunskill teaches courses in reinforcement learning, probabilistic modeling, and decision-making under uncertainty, contributing to curricula that align with programs at Stanford University, Massachusetts Institute of Technology, and graduate training supported by the National Science Foundation Graduate Research Fellowship Program. Her mentees have gone on to positions at academic institutions including Carnegie Mellon University, University of California, Berkeley, and industry research groups at DeepMind and OpenAI. She serves on program committees for conferences such as NeurIPS, ICML, and AAAI, and contributes to organizing workshops that connect researchers from University of Oxford, ETH Zurich, and University of Cambridge.

Category:Computer scientists Category:Reinforcement learning researchers