Taylor A. Kessler Faulkner

The University of Texas at Austin, University of Washington

Papers

Total Citations

H-Index

About

Taylor A. Kessler Faulkner is a leading researcher at the intersection of robotics, human-robot interaction, and machine learning, with a primary focus on enabling robots to learn effectively from imperfect human teachers. Her core research addresses a critical challenge in interactive reinforcement learning (RL): how robots can leverage feedback from people—who may be inattentive, misunderstand the task, or provide noisy advice—without compromising learning performance. In her highly cited 2020 work on interactive RL with inaccurate feedback (14 citations), she pioneered algorithms that allow agents to learn from both environmental rewards and unreliable human input. Her 2021 study on learning from imperfect teachers (9 citations) further advanced this area by developing methods for robots to capitalize on feedback only when teachers are attentive. Beyond learning algorithms, Faulkner has made notable contributions to assistive robotics, co-designing and evaluating a robot-assisted feeding system for out-of-lab use with real users. This work, published in 2025, demonstrates her commitment to translating algorithmic innovations into tangible systems that empower people with motor impairments to eat independently, addressing a deeply human need. Her research is widely recognized for its rigor and real-world impact.