software

libPG -- A high performance distributed RL library.

This is a high-speed C++ implementation of many popular RL algorithms for MDPs and POMDPs including:

Baxter and Bartlett's GPOMDP/OLPOMDP
Jan Peter's Natural Actor-Critic
Various PSR code from Satinder Singh's papers
Online PSRs from McCracken and Bowling
HMM estimation of hidden state from observations
Finite history methods
Optimised for multi-agent/distributed RL applications

Requires the uBlas components of the Boost library. Having Lapack and Atlas will also open up more features.

This is the source code for a few papers, including the base library for anything to do with the Factored Policy Gradient Planner (FPG), the Brazil Planner, and the RL with PSRs/POMDPs paper.

FPG (version IPC-06)

[link] - [SH link]

Probabilistic planner addressing its problem as a reinforcement learning one. Its principle is to optimise a controller's parameters (such as a neural network whose input is a state and whose output is a decision) using a domain simulator. It relies on the libPG library (see above).

FPG's first version is meant to deal with concurrent temporal probabilistic planning. It can even handle uncertain continuous durations and resources.
Another version (FPG-ipc) entered in the probabilistic track of the fifth international planning competition (IPC-5). It is based on the MDPSim simulator.
Rank: 1^st (behind FF-replan).
Note: "FPG" (Factored Policy-Gradient) is a name that also covers an algorithm optimizing traffic control (through traffic lights).