Piqle (Platform for Implementing Q-Learning Experiments) is a Java framework for fast design, prototyping and test of reinforcement learning experiments (RL).
By clearly separating algorithms and problems, it allows users to focus on either part of the RL paradigm:designing new algorithms or implementing new problems.
Piqle implements many classical RL algorithms, making their parameters easily tunable.
At this time, 13 problems are implemented, several with one or more variants. The user's manual explains in detail how to code a new problem.
Written in Java, Piqle is as platform-independent as Java itself. Its components can easily be embedded as part of complex implementations, like robotics or decision making.
Using Piqle's components makes the definition and test of algorithms experimentation, comparison or benchmarking straightforward for Java programmers.
A rather complete 93 pages user's manual, together with an up-to-date javadoc, are provided.
Core program, examples and documentation are all available under the GNU Lesser General Public License at Sourceforge.
Piqle development is active, new features and implementations in test can be downloaded in the project's Subversion repository