Reinforcement learning is one of the most active research areas in artificial intelligence and applies to a wide range of use cases in different sectors. What makes the technology unique in that it creates autonomous systems that learn from trial-and-error interaction to maximise the total amount of reward it receives while interacting with a complex, uncertain environment.
To provide students with the skills needed in a transforming AI landscape, the ENSI school (National School for Computer Science) invited us to give a training course on the subject. Last week, our research engineer Nourchène gave a 15-hour module aimed at final year engineering students in the Software Engineering programme and the Intelligent Systems master.
During the course, she gave an introduction to reinforcement learning and its main principles :
- What makes it different from traditional machine learning techniques (supervised and unsupervised ML)
- Modelling by Markovian processes
- The three main families of RL algorithms
- The RL taxonomy with examples of algorithms from each family
- A practical part with the OpenAI Gym toolkit to familiarise students with the different environments available and test some of them.
If you wish to know more about RL, do not hesitate to reach out to research.euranova.eu!
Thank you to the ENSI team for the invitation. It was a pleasure for us to be able to exchange with students. If you are a student interested in the field of Reinforcement Learning, we propose graduation projects on the subject. You can find all the internship offers on our website.