Talk at the MIT Embodied Intelligence Seminar (Feb 2021):

Claudia Pérez D'Arpino

I believe robotics research offers a unique window into discovering the principles of intelligence in interaction with people and the world. My research focuses on human-machine hybrid learning as a framework for developing intelligent agents that can learn from observations, augment this learning with self-exploration, and apply the learned skills in interaction with humans.

I am a Postdoctoral Scholar in Computer Science at Stanford University, Stanford Vision and Learning Lab (SVL). I work on robot learning for collaborative tasks in manipulation and social navigation.

I received my PhD in Computer Science at MIT (2019) advised by Prof. Julie Shah in the Interactive Robotics Group at CSAIL. My PhD thesis focused on Hybrid Learning for Multi-Step Manipulation in Collaborative Robotics. I participated in the DARPA Robotics Challenge (2012-2015) with Team MIT.

Previously I was an Assistant Professor in the Electronics and Circuits Department at the Simón Bolívar University, where I also received my Master and Engineering degrees in Electronics Engineering.

Google Scholar -- @cpdarobotics -- Youtube -- Linkedin


- Social Navigation Challenge at CVPR 2021. [Challenge Website] [EmbodiedAI Workshop] [Blog post]
- ICRA 2021: Robot Navigation in Constrained Pedestrian Environments using Reinforcement Learning. [Project] [Paper]
- Co-teaching a new course at Stanford: CS331B: Interactive Simulation for Robot Learning
- New release of the iGibson simulator: [Project] [Preprint]
- Preprint: Experimental assessment of human-robot teaming for multi-step remote manipulation with expert operators. [Project] [Preprint]