Jocelyn Shen

About Me

Hi! I'm Jocelyn, a Master's student at the MIT Media Lab's Personal Robots Group. I graduated from MIT in 2021 with a degree in Computer Science alongisde a minor in Economics. My research interests are at the intersection of human-robot/human-computer interaction, behavioral science, and NLP. In the past, I've worked at Facebook (Software Engineering Intern), Citadel (Software Engineering Intern), and Affectiva (Data Science Intern). Aside from my academic work, I like to sketch characters and write anything ranging from poems to essays to novels. Check out the rest of this site to learn more!

Research

Lifelong Personalization via Gaussian Process Modeling for Long-Term HRI

Samuel Spaulding, Jocelyn Shen, HaeWon Park, and Cynthia Breazeal. Frontiers in Robotics and AI, 2021

Towards Transferrable Player Models in Educational Games

Samuel Spaulding, Jocelyn Shen, HaeWon Park, and Cynthia Breazeal. Proceedings of the 20th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2021

Branching principles of animal and plant networks identified by combining extensive data, machine learning and modelling

Alexander B. Brummer, Panagiotis Lymperopoulos, Jocelyn Shen, Elif Tekin, Lisa P. Bentley, Vanessa Buzzard, Andrew Gray, Imma Oliveras, Brian J. Enquist and Van M. Savage. Journal of the Royal Society Interface Volume 18, Issue 174, 2021

Projects

Role of Parent Nonverbal Cues in Children’s Engagement During Dialogic Reading —MAS.630 Affective Computing, Fall 2021

Research paper exploring the effect of a parent's nonverbal cues (body gesture) on a child's affect (engagement) during a joint educational task (dialogic reading).

Improving Multiturn Dialogue Reasoning Through Pretraining —6.864 Advanced NLP, Fall 2021

Evalute BERT-based models' abilities to perform various types of reasoning in multiturn dialogue datasets using different pretraining approaches.

pop'n —Weblab 2020 Semifinalist

Online polling game to test people's knowledge on questions from around the world

Eye in the Sky —HackMIT Winner 2019 (Health Tech)

Mobile application for crowd-sourced natural disaster relief.

footprnt —Facebook, 2019

Travel the world virtually through footprnt. Explore user-created travel logs from all around the world.

In-Vehicle Object Detection —Affectiva, Emotion AI Summit 2018

Object detection using Retinanet to help drivers on the road.