We have an open PhD position on the intersection of Reinforcement Learning and Control theory. The Ph.D. will be conducted within Automatic Control division at Linköping University, Sweden. The candidate will be jointly supervised by Farnaz Adib Yaghmaie and Gustaf Hendeby.
Project Description: Reinforcement Learning (RL) provides an end-to-end framework to control dynamical systems, from sensory data to control input, and has shown astonishing performance for example in playing strategic games and surpassed human champion performance. This project aims to study, analyze, and develop RL algorithms for control of partially observable dynamical systems.
- To study and quantify the effect of partial observability, noise, and uncertainty
- To study how data can be efficiently used in RL algorithms including model-free and model-building approaches
- To develop efficient algorithms for partially observable dynamical systems
Qualifications: We are looking for an enthusiastic and ambitious candidate with the following qualifications
- Solid background in Control Theory
- Knowledge of Reinforcement Learning
- Solid programming skills
- Great communication skills and proficiency in English
It is meriting to have knowledge in Machine Learning, optimization, and statistics. Knowledge of programming with Python is also an advantage.
Location: Linköping University, Sweden
Salary: The starting salary is 32700 SEK per month before tax.
Starting time: as soon as possible
Deadline to apply: Dec 05.
Link: https://liu.se/en/work-at-liu/vacancies?rmpage=job&rmjob=20436&rmlang=UK
