Scholar
Joe Watson
Google Scholar ID: xLtXIZAAAAAJ
University of Oxford
Citations & Impact
Citations
423
H-index
11
i10-index
15
Publications
20
Co-authors
14
list available
Publications
Resume (English only)
Academic Achievements
- 1. 'Inferring Smooth Control: Monte Carlo Posterior Policy Iteration with Gaussian Processes', presented at the 2022 Conference on Robot Learning (oral); 2. 'Coherent Soft Imitation Learning', presented at the 2023 Advances in Neural Information Processing Systems (NeurIPS) [spotlight]; 3. 'Stochastic Control as Approximate Input Inference', under review; 4. 'Neural Linear Models with Gaussian Process Priors', presented at the 2021 Advances in Approximate Bayesian Inference (AABI), co-first author with J. A. Lin; 5. 'Latent Derivative Bayesian Last Layer Networks', presented at the 2021 Artificial Intelligence and Statistics (AISTATS), co-first author with J. A. Lin; 6. 'A Differentiable Newton-Euler Algorithm for Real-World Robotics', submitted to IEEE Transactions on Robotics; 7. 'Benchmarking Structured Policies and Policy Optimization for Real-World Dexterous Object Manipulation', published in the IEEE Robotics and Automation Letters, Special Issue: Robotic Grasping and Manipulation Challenges and Progress.
Research Experience
- 1. Postdoctoral researcher at the Applied Artificial Intelligence Lab, Oxford Robotics Institute, University of Oxford, researching world models and robot learning methods; 2. Before starting his PhD, worked on developing Versius, a novel robotic system for laparoscopic surgery, from prototype to product; 3. Internship at Google DeepMind with the Robotics team, hosted by Sandy Huang and Nicholas Heess, during 2022-23.
Education
- PhD: 2024 from TU Darmstadt, supervised by Prof. Jan Peters; Bachelor's: Information Engineering at Peterhouse, University of Cambridge, awarded the Charles Babbage senior scholarship.
Background
- Research Interests: robotics, optimal control, approximate inference, system identification. Professional Field: robot and machine learning. Introduction: Currently a postdoctoral research assistant at the Oxford Robotics Institute, University of Oxford, working on world models and robot learning methods for sensorimotor manipulation.
Miscellany
- Further Links: researchgate, mendeley, linkedin
