Ria Doshi

Hi! I’m a fourth-year undergraduate student at UC Berkeley studying computer science. My research focuses on building generalist robot policies, through cross-embodied data, vision-language pre-training, and autonomous policy learning. I’ve had the pleasure of working with Professor Sergey Levine and Professor Jeannette Bohg on the above. Aside from robot learning, I’ve done research in emergent phenomena of LLMs with Professor Anant Sahai. I’m also the author of The Code Detectives, a book series about young girls using AI to solve mysteries.

Email      Google Scholar      LinkedIn      Twitter

Scaling Cross-Embodied Learning: One Policy for Manipulation, Navigation, Locomotion and Aviation

Ria Doshi*, Homer Walke*, Oier Mees, Sudeep Dasari, Sergey Levine
(Oral Presentation) Conference on Robot Learning (CoRL) 2024
Webpage  •   PDF  •   Code

Shadow: Leveraging Segmentation Masks for Cross-Embodiment Policy Transfer

Marion Lepert, Ria Doshi, Jeannette Bohg
Conference on Robot Learning (CoRL) 2024
Webpage  •   PDF

Dexterous Manipulation from Images: Autonomous Real-World RL via Substep Guidance

Kelvin Xu*, Zheyuan Hu*, Ria Doshi, Aaron Rovinsky, Abhishek Gupta, Vikash Kumar, Sergey Levine
International Conference on Robotics & Automation (ICRA) 2022
Webpage  •   PDF

PolyGround: Towards a Playground for Large Language Model Alignment & PEFT Using Polynomial Regression

Ria Doshi*, Morten Svendgård*, Max Wilcoxson*, Dylan Davis*, Reya Vir, Anant Sahai
International Conference on Machine Learning (ICML) Workshop 2024
Webpage  •   PDF

Octo: An Open-Source Generalist Robot Policy

Dibya Ghosh*, Homer Walke*, Karl Pertsch*, Kevin Black*, Oier Mees*, Sudeep Dasari, Joey Hejna, Tobias Kreiman, Ria Doshi, Charles Xu, Jianlan Luo, You Liang Tan, Lawrence Yunliang Chen, Pannag Sanketi, Quan Vuong, Ted Xiao, Dorsa Sadigh, Chelsea Finn, Sergey Levine
Robotics, Sciences & Systems (RSS) 2024
PDF  •   Code


Open X-Embodiment: Robotic Learning Datasets and RT-X Models

Open X Embodiment Collaboration et. al.
International Conference on Robotics & Automation (ICRA) 2023
Webpage  •   PDF

ML Research Intern, Wayve AI

My work focused on building LINGO, towards a vision-language-action model for autonomous driving. I was involved with building a large-scale inference pipeline using Kubernetes, improving road-signal perception capabilities, and data curation for referential segmentation.
Project Webpage

ML Consultant, Spyne AI

I developed high-fidelity, 3D renderings of cars using NERF models for smarter, more efficient online marketing. Implemented interpolation & feature matching techniques for output consistency across varying input dimensions.

Author, The Code Detectives Book Series

I've always been a big fan of mystery novels, especially when I was younger. I really enjoyed writing The Code Detectives, about two young girls who use coding & AI to solve neighborhood mysteries. Writing the books has been one of the most gratifying projects I've worked on.
Book 1  •   Book 2

2024

Oral Presentation, Conference on Robot Learning

2024

Invited Talk, Google DeepMind

2024

Invited Talk, Wayve AI