I study human and machine learning in socially interactive tasks with humans and artificial agents
I am a Ph.D. student in the Personal Robots Group at the MIT Media Lab, advised by Professor Cynthia Breazeal. My research broadly focuses on designing, programming, and evaluating robotic systems to create engaging and natural Human-Robot Interaction.
My work aims to advance AI and Robotics by building intelligent systems that can successfully interact with others to achieve goals in the real world. In addition to other robots and intelligent agents, real world robots are increasingly interacting closely and in-person with humans, so an important component of my research focuses on building robots that can interact flexibly and fluently with humans.
My thesis research has focused on designing social robots that can construct models of human interaction partners from diverse sources of multimodal data, take actions based on these models, and reason about and refine its actions and the models of the interaction partner(s). Specifically, my Ph.D. work introduces and evaluates a new class of “lifelong personalization” models that are transferrable and continual – that is, they personalize to partners across tasks and over time. I believe that AI and Robotics are transformative technologies and have a special passion for finding creative, ethical, and impactful applications for these fields (especially in Education and Art).
From 2015-2017, I worked at Jibo Inc. on the Robot Skills and Character AI teams, helping to develop the world’s first social robot for the home. Before coming to MIT, I received a B.S. in Computer Science from Yale University. While there, I worked with Professor Brian Scassellati in the Yale Social Robotics Lab and spent summers working at Amazon.com and at Disney Research with Senior Research Scientist Jonathan Yedidia.
Connect with me on Twitter @SpauldingSam
I gave a Spotlight Talk on my work to develop transferrable models of student learning at the RL4ED Workshop held at EDM 2021
Paper Accepted! Lifelong Personalization via Gaussian Process Modeling for Long-term HRI was published in the latest issue of Frontiers in Robotics & AI
I am Co-organizing a workshop on Lifelong Learning and Personalization in Long-Term Human-Robot Interaction (LEAP-HRI) at HRI 2021
Paper Accepted! Towards Transferrable Personalized Student Models in Educational Games' was accepted as a full paper at AAMAS 2021.
I gave a talk (by video) at the UCHicago HCI Club Seminar series. Thanks to the organizers for inviting me!