PhD - Scalable and Efficient Reinforcement Learning Methods for Physical AI (f/m/div.)
Core
Researching scalable and efficient reinforcement learning methods to develop resource-efficient autonomous decision-making systems for real-world physical applications like automated driving and robotics.
Role type
PhD researcher in Reinforcement Learning and Physical AI
Builds
AI agents capable of continuous improvement, adaptation to novel situations, and effective simulation-to-reality transfer
Domain
Robotics, Autonomous Systems, Machine Learning
Deliverable
production ML models
Required skills
Reinforcement Learning, Imitation Learning, Safe RL, Offline RL, Multi-agent RL, Python, PyTorch, JAX, TensorFlow, C++, ROS 2, MLOps, CI/CD
Preferred skills
Large-scale simulation, World models, Model compression, Generative AI, Top-tier conference publications (NeurIPS, ICML, ICLR, CVPR, IROS, ICRA)
Responsibilities
Develop efficient autonomous decision-making systems, combine RL with simulation and world models, optimize for resource efficiency and generalization, collaborate with interdisciplinary teams
Seniority
PhD Candidate
