Scholar
Sumedh Sontakke
Google Scholar ID: sUb_eyUAAAAJ
Frontier AI and Robotics, Amazon
Citations & Impact
Citations
2,097
H-index
10
i10-index
11
Publications
20
Co-authors
6
list available
Publications
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
- Published multiple papers such as 'Video2Skill: Adapting Events in Demonstration Videos to Skills in an Environment using Cyclic MDP Homomorphisms' and 'Model2Detector: Widening the Information Bottleneck for Out-of-Distribution Detection using a Handful of Gradient Steps.' Involved in projects like RoboCLIP and RT-1, where RT-1 was a Best Demo Paper Finalist at Robotics Science and Systems 2023.
Research Experience
- Worked at Facebook-funded Skyline Labs; conducted research at the Adolphs Lab at Caltech, University of Oxford (as a SENS Scholar), and Max Planck Institute for Intelligent Systems in Tübingen, Germany; interned at Google Brain (now Google Deepmind), Bell Labs, Adobe Research, and the Robot Learning Group at Microsoft Research.
Education
- Received his PhD in 2024 from USC, advised by Prof Laurent Itti and Prof Erdem Bıyık; Bachelor's in Electrical Engineering from the College of Engineering, Pune in 2019.
Background
- An Applied Scientist at Amazon Robotics, with research interests in autonomous decision-making, causality, and physical artificial intelligence. His doctoral thesis investigated foundation models for embodied AI.
Miscellany
- Reviewer for ICLR, AISTATS, ICML, NeurIPS, and CoRL.
