Ashwin Balakrishna
Scholar

Ashwin Balakrishna

Google Scholar ID: tfN6V84AAAAJ
Physical Intelligence
RoboticsMachine LearningReinforcement LearningImitation Learning
Citations & Impact
All-time
Citations
4,734
 
H-index
28
 
i10-index
38
 
Publications
20
 
Co-authors
9
list available
Resume (English only)
Academic Achievements
  • Selected Publications: 'Gemini Robotics: Bringing AI into the Physical World', 'A Taxonomy for Evaluating Generalist Robot Policies', 'Robot Data Curation with Mutual Information Estimators', 'Robo-DM: Efficient Robot Big Data Management', 'GHIL-Glue: Hierarchical Control with Filtered Subgoal Images', 'OpenVLA: An Open-Source Vision-Language-Action Model'; Awards: Best Robot Learning Paper at ICRA, Outstanding Paper Award Finalist at CoRL.
Research Experience
  • Work Experience: Research Scientist at Physical Intelligence; Research Projects: Gemini Robotics, A framework and benchmark to evaluate the generalization capability of robotic manipulation policies, Filtering robotic manipulation datasets for high quality demonstrations, Robo-DM, GHIL-Glue, OpenVLA, etc.
Education
  • PhD in Computer Science at UC Berkeley's AUTOLAB; Bachelor's degree in Electrical Engineering at Caltech.
Background
  • Research Interests: Building generally intelligent robots; Professional Field: Computer Science; Background: A Research Scientist at Physical Intelligence (previously Google DeepMind), focusing on bridging the general-purpose capabilities of vision and language foundation models with algorithmic paradigms for learning from interaction, such as imitation learning and reinforcement learning.
Miscellany
  • Contact Information: Email, Resume, CV, Google Scholar, Twitter, LinkedIn.