Gautam Salhotra

Gautam Salhotra

Robotics & AI Research

Google DeepMind

I am a Research Scientist at Google DeepMind (GDM). Previously, I was a robotics researcher at Intrinsic AI, where I worked with Stefan Schaal. I completed my PhD in Computer Science at the University of Southern California (USC), advised by Gaurav Sukhatme.

My research focuses on foundation models for robotics and Reinforcement Learning (RL). My doctoral thesis explored learning visuomotor manipulation skills for robotic arms, utilizing Learning from Demonstration (LfD) and RL. I have applied these methods to complex challenges such as dexterous cloth manipulation and precision assembly tasks. More broadly, I am interested in achieving greater autonomy for robots by combining planning algorithms with learning, and developing scalable learning frameworks for physical agents.

Interests
  • Dexterous Robot Manipulation
  • Learning from Demonstrations
  • Reinforcement Learning
Education

  • PhD in Computer Science (Robotics), 2024

    University of Southern California

News

Selected Publications

See Google Scholar for a full list

Shashank Hegde, Satyajeet Das, Gautam Salhotra, Gaurav S. Sukhatme (2025). Latent Weight Diffusion: Generating reactive policies instead of trajectories.

Cite PDF

Open X-Embodiment Collaboration, (multiple authors), Gautam Salhotra (2024). Open X-Embodiment: Robotic Learning Datasets and RT-X Models : Open X-Embodiment Collaboration. In ICRA 2024, Best Paper.

Cite Website PDF Code

Gautam Salhotra, I-Chun Arthur Liu, Gaurav S. Sukhatme (2023). Learning Robot Manipulation from Cross-Morphology Demonstration. In CoRL 2023.

Cite Website PDF Code

Gautam Salhotra, I-Chun Arthur Liu, Marcus Dominguez-Kuhne, Gaurav S. Sukhatme (2022). Learning Deformable Object Manipulation from Expert Demonstrations. In IEEE RA-L, IROS ‘22.

Cite Website PDF Code

Jun Yamada, Youngwoon Lee, Gautam Salhotra, Karl Pertsch, Joseph J. Lim, Peter Englert (2020). Motion Planner Augmented Reinforcement Learning for Robot Manipulation in Obstructed Environments. In CoRL ‘20.

Cite Website PDF Code

Contact