Brad Knox
Research Associate Professor
Brad Knox is a Research Associate Professor with a diverse research background encompassing machine learning, human-computer interaction, and computational models of human behavior for cognitive science research. His research primarily focuses on the human side of reinforcement learning, including pioneering work on human-in-the-loop reinforcement learning that earned him the 2012 best dissertation award for the UT Austin Department of Computer Science.
Research
Select Publications
S Amershi, M Cakmak, WB Knox, T Kulesza. 2014. Power to the people: The role of humans in interactive machine learning. AI Magazine Vol. 35.
S Booth, BW Knox, J Shah, S Niekum, P Stone, A Allievi. 2023. The perils of trial-and-error reward design: misdesign through overfitting and invalid task specifications.
Toward Believable Acting for Autonomous Animated Characters. 2022. Toward Believable Acting for Autonomous Animated Characters.
WB Knox, S Hatgis-Kessell, S Booth, S Niekum, P Stone, A Allievi. 2022. Models of human preference for learning reward functions.
WB Knox, A Allievi, H Banzhaf, F Schmitt, P Stone. 2021. Reward (Mis)design for Autonomous Driving.
Awards & Honors
2018 -
Hasbro Emerging Innovator Award (Finalist; lead technologist on the project for Dash Robotics)
2016 -
Awarded NSF Small Business Innovation Research (SBIR) grant as PI
2013 -
AI 10 to Watch by IEEE Intelligent Systems
2013 -
Bert Kay Dissertation Award (for best dissertation from UT Austin Computer Science)
2013 -
IFAAMAS-12 Victor Lesser Distinguished Dissertation Award (Runner-up)
2013 -
ICSR Best Paper Award
2012 -
Ro-Man CoTeSys Cognitive Robotics Best Paper (Finalist)
2010 -
AAMAS Pragnesh Jay Modi Best Student Paper Award
2008 to 2011 -
NSF Graduate Research Fellowship
