Introduction
Reinforcement Learning (RL) is a class of machine learning algorithms in which an agent interacts by trial-and-error in an environment.
RL in conjunction with Deep Learning has obtained outstanding results in Atari video games, the Go board-game and a more complex environment like StarCraft II.
Recently many open source RL frameworks has been released by software companies in order to easily train and test new RL algorithms.
The goal of the thesis is to benchmark the most promising RL frameworks, to study the new algorithms proposed and to evaluate their performance on research environments.

Planned Activities
- Acquire strong theoretical basis on Deep Reinforcement Learning;
- Install and compare the different RL frameworks;
- Adapt and apply the best framework to a real application.
Required Skills:
- Experience with Linux or Unix based OS;
- Proficiency in at least one programming language (Python, Lua, Matlab, C++, Java);
- Basic knowledge of machine learning;
- Good knowledge of linear algebra.
Competencies to be acquired:
- Experience with the application of Machine Learning to complex systems.
- Expertise on the most recent Deep Reinforcement Learning algorithms;
- Proficient use of the most advanced development frameworks with a software engineering approach.
Who we’re looking for
Students that are about to get their Master Degree in: computer science, computer engineering, mechatronic engineering, mathematical engineering, physics of complex systems, mathematics, physics or stochastics and data science.
Duration of this Projects: 5-6 months
Check these links before moving on
[1] Reinforcement Learning: An Introduction
http://incompleteideas.net/book/the-book.html
[2] Deep-Q-Network
https://www.nature.com/articles/nature14236
[3] AlphaGo
https://deepmind.com/research/alphago/
[4] AlphaStar https://bit.ly/2B5YrKh
[5] RL frameworks:
https://github.com/deepmind/trfl
https://github.com/facebookresearch/Horizon
https://github.com/google/dopamine
https://github.com/NervanaSystems/coach
https://github.com/openai/spinningup
[6] OpenAI gym
[7] Mujoco
[8] CARLA Simulator
Contact Us
Directly by email to: [email protected]
By LinkedIn: linkedin.com/in/cannavò-sonia-66a95467