  Peter Karkus  

 



  ![](/sites/default/files/person/Karkus.jpg)

  

 Peter is a Research Scientist at NVIDIA. Previously he has been a PhD candidate at the National University of Singapore, and he also held visiting research appointments at MIT and CMU.

Peter's research vision is to build human-level robot intelligence by combining structure and learning. His interests cover robotics, machine learning, and autonomous vehicles. His recent works are on neural networks that encode differentiable robot algorithms in order to learn partially observable planning, visual navigation, mapping and localization tasks.



   Research Area(s)

[Artificial Intelligence and Machine Learning ](/index.php/research-area/machine-learning-artificial-intelligence)

[Autonomous Vehicles](/index.php/research-area/autonomous-vehicles)

[Robotics](/index.php/research-area/robotics)

 

 

  

 Main Field of Interest

[Autonomous Vehicles](/index.php/research-area/autonomous-vehicles)

 

  

 Google Scholar

[https://scholar.google.com/citations?user=cjUid0YAAAAJ&amp;hl=en&amp;oi=ao](https://scholar.google.com/citations?user=cjUid0YAAAAJ&hl=en&oi=ao)

 

  

 

 

 



 ### Publications

 

### 2025 

[Beyond Behavior Cloning in Autonomous Driving: a Survey of Closed-Loop Training Techniques](/publication/2025-12_beyond-behavior-cloning-autonomous-driving-survey-closed-loop-training)

[Peter Karkus](/person/peter-karkus), [Maximilian Igl](/person/maximilian-igl), [Yuxiao Chen](/person/yuxiao-chen), Kashyap Chitta, Jef Packer, [Bertrand Douillard](/person/bertrand-douillard), [Thomas Tian](/person/thomas-tian), Alexander Naumann, Guillermo Garcia-Cobo, Shuhan Tan, [Alperen Degirmenci](/person/alperen-degirmenci), Alexander Popov, Nikolai Smolyanskiy, Urs Muller, [Boris Ivanovic](/person/boris-ivanovic), [Marco Pavone](/person/marco-pavone)













### 2023 

[Planning with Occluded Traffic Agents using Bi-Level Variational Occlusion Models](/publication/2023-05_planning-occluded-traffic-agents-using-bi-level-variational-occlusion-models)

Filippos Christianos, [Peter Karkus](/person/peter-karkus), [Boris Ivanovic](/person/boris-ivanovic), Stefano V. Albrecht, [Marco Pavone](/person/marco-pavone)



[IEEE International Conference on Robotics and Automation (ICRA) 2023](https://www.icra2023.org/)









[Tree-structured Policy Planning with Learned Behavior Models](/publication/2023-05_tree-structured-policy-planning-learned-behavior-models)

[Yuxiao Chen](/person/yuxiao-chen), [Peter Karkus](/person/peter-karkus), [Boris Ivanovic](/person/boris-ivanovic), [Xinshuo Weng](/person/xinshuo-weng), [Marco Pavone](/person/marco-pavone)



[IEEE International Conference on Robotics and Automation (ICRA) 2023](https://www.icra2023.org/)









### 2022 

[DiffStack: A Differentiable and Modular Control Stack for Autonomous Vehicles](/publication/2022-12_diffstack-differentiable-and-modular-control-stack-autonomous-vehicles)

[Peter Karkus](/person/peter-karkus), [Boris Ivanovic](/person/boris-ivanovic), [Shie Mannor](/person/shie-mannor), [Marco Pavone](/person/marco-pavone)



[Conference on Robot Learning (CoRL) 2022](https://corl2022.org/)