Machine Learning methods for autonomous adaptation of Human-Machine Interaction to the unique needs and preferences of users over time
Co-advised by professors Maja Matarić and Fei Sha, I was dually trained in the fields of Machine Learning and Artificial Intelligence as well as Human-Computer Interaction and Human-Machine Interaction.
Specifically, my graduate research developed and studied computational methods to adapt socially assistive robots to meet the needs of individual users over time.
I am deeply connected to issues surrounding neurodiversity and disability rights, so much of my work has focused on applications to differentiated education and personalized health care. At USC, I was honored by the 2018 Viterbi Undergraduate Research Mentoring Award and The Order of Areté for mentoring and K-12 STEM outreach activities.
Socially Assistive Robotics (SAR) has potential to supplement expensive and often inaccessible therapy for children affected with Autism Spectrum Disorder (ASD). As part of this multi-university NSF Expedition in Computing, we are deploying SAR systems in the family homes of children with ASD for 30 day interventions. The intervention focuses on sequencing and ordering, turn taking, and emotion recognition skills, adapting its instruction to each individual child over time and socially engaging them in learning through play.
In this research, we developed a Socially Assistive Robot (SAR) tutoring system to support the efforts of educators in teaching number concepts to preschool children. The SAR system was iteratively designed with the input of education experts to be developmentally appropriate. The system was deployed in a real-world preschool classroom. The data collected were used to train personalized models of number concepts learning, leveraging the multimodal data, domain knowledge, and learning style.
Caitlyn Clabaugh and Maja Matarić, "Escaping Oz: Autonomy in Socially Assistive Robotics" in Annual Review of Control, Robotics, and Autonomous Systems, vol. 2, 2019.
Caitlyn Clabaugh, "Robot Learning," in The Robotics Primer (2nd Edition), Maja Matarić, MIT Press, 2019.
Caitlyn Clabaugh and Maja Matarić, "Can Technology Save Children from Itself?," in Topical Issue on Roboethics, Paladyn. Journal of Behavioral Robotics, 2018.
Caitlyn Clabaugh and Maja Matarić, "Robots for the People, by the People: Personalizing Human-Machine Interaction," in Science Robotics, vol. 3, no. 21, p.eaat7451, 2018. PDF
Caitlyn Clabaugh, David Becerra, Eric Deng, Gisele Ragusa, and Maja Matarić, "Month-long, In-home Case Study of a Socially Assistive Robot for Children with Autism Spectrum Disorder," in Companion of the Twenty-Second ACM/IEEE International Conference on Human-Robot Interaction (HRI), Chicago, IL, March 5-8, 2018. PDF
Caitlyn Clabaugh and Maja Matarić, "Exploring Elicitation Frequency of Learning-Sensitive Information by a Robotic Tutor for Interactive Personalization," in Twenty-Fifth ACM/IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), New York, NY, August 26-31, 2016. PDF Best Empirical Paper
Caitlyn Clabaugh, Gisele Ragusa, Fei Sha, and Maja Matarić, "Designing a Socially Assistive Robot for Personalized Number Concepts Learning in Preschool Children," in Fifth International Conference on Development and Learning and on Epigenetic Robotics (ICDL-EPIROB), Providence, RI, August 13-16, 2015. PDF
Caitlyn Clabaugh, Konstantinos Tsiakas, and Maja Matarić, "Predicting Preschool Mathematics Performance of Children with a Socially Assistive Robot Tutor," in Workshop on Synergies Between Learning and Interaction at the 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2017), Vancouver, Canada, September 28, 2017. PDF
Caitlyn Clabaugh, "Interactive Personalization for Socially Assistive Robotics," in HRI-Pioneers Workshop at the Twenty-First ACM/IEEE International Conference on Human-Robot Interaction (HRI), Vienna, Austria, March 6-9, 2017. PDF
Braden McDorman, Caitlyn Clabaugh, and Maja Matarić, "Attachment Theory in Long-Term Human-Robot Interaction," in Workshop on Long-Term Child-Robot Interaction held at the Twenty-Fifth ACM/IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN), New York, NY, August 31, 2016. PDF
Caitlyn Clabaugh, Tejas Ram, and Maja Matarić, "Estimating Visual Focus of Attention in Dyadic Human-Robot Interaction for Planar Tasks," in First International Workshop on Educational Robots (WONDER) at the Seventh International Conference on Social Robotics (ICSR), Paris, France, October 26, 2015. PDF
Caitlyn Clabaugh, Fei Sha, Gisele Ragusa, and Maja Matarić, "Towards a Personalized Model of Number Concepts Learning in Preschool Children," in Workshop on Machine Learning for Social Robotics at ICRA, Seattle, WA, May 26, 2015. PDF
Jillian Greczek, Elaine Short, Caitlyn Clabaugh, Katelyn Swift-Spong, and Maja Matarić, "Socially Assistive Robotics for Personalized Education for Children," in AAAI Fall Symposium Series, Washington, DC, November 13-15, 2014. PDF
Ross Mead, Amin Atrash, Edward Kaszubski, Aaron St Clair, Jillian Greczel, Caitlyn Clabaugh, Brian Kohan, and Maja Matarić, "Building Blocks of Social Intelligence: Enabling Autonomy for Socially Intelligent and Assistive Robots," in AAAI Fall Symposium Series, Washington, DC, November 13-15, 2014. PDF
Elizabeth Lee, "Robots Could Be Girls' Ticket to Future," Research Featured in Voice of America (VOA), April 18, 2018. Online
Marc Ballon, "Marooned Robots Love Math," Research Featured in USC Viterbi Magazine, Fall 2016. Online
I've been fortunate to mentor over 40 undergraduate and graduate students at USC, and help lead STEM outreach activities and after-school programs that have reached thousands of K-12 students.
I helped pioneer and develop curriculum for the USC Robotics & Coding Academy after-school program that gives USC students the opportunity to teach robotics to 70 local middle schoolers, annually.
I've led and participated in numerous robotics and STEM workshops and programs for K-12 students, including All Star Code, the Global Conference on Educational Robotics and Botball, USC's SHINE program, and many school assemblies.
|Applications||Unity, MATLAB, ROS, Vim, Emacs, R, Visual Studios, .NET Framework, Eclipse, NetBeans, OpenGL, MS Office, LaTeX, Adobe Photoshop, Illustrator, Premiere, After Effects and Dreamweaver, Maya, Mudbox, Final Cut Pro|
|Operating Systems||OSX, Linux/Unix, Android, Windows 7, Vista, and XP|
|Robotic Platforms||USC's SPRITE robot, Aldebaran NAO, MIT Media Lab's DragonBot, Wonder Workshop Dot & Dash, Sphero|
|Natural Languages||English (native speaker), French (limited working proficiency), Spanish (elementary proficiency)|
|Citizenship||United States of America|
|Other||Assistant Cake Decorator (2015-2016), USC Boxing Team (2014-2016)|