The project began in the late 2004 and since then it has been developed and mantained by Marcello Restelli, Alessandro Lazaric and Enrique Munoz de Cote under the supervision of prof. Andrea Bonarini.
PRLT is a framework for the development of Reinforcement Learning algorithms and is a support for experimental activities both in Single and Multiagent problems. Many other systems are available for Reinforcement Learning (see a complete list on the Learning and Artificial Intelligence Group website at the University of Alberta) and with PRLT we would like to contribute to the development of complex tools in order to push Reinforcement Learning towards more and more complex problems and applications. Furthermore, we contributed to the environment repository of the RL-Glue Library and we participated to the benchmark events at NIPS 2005 and ICML 2006.
Many students significantly contributed to the development of the toolkit and we want to thank them all for their work, suggestion and passion.
Here you can find a list of all the people currently involved in the development of PRLT and students who contributed in the past with references to their master theses.