This is a third year computerscience project.
A software system for simulating and animating Reinforcement Learning (RL) algorithms mainly for modular robots.
Finds bad examples for the local search Path Partitioning algorithm
Path Partitioner constructs random graph partitions and solves the linear (or integer) programming problems that assign weights to the edges such that the performance ratio of the local search algorithm for the Path Partitioning problem is minimized. See http://www.tau.ac.il/~ohadschn/Thesis/Partitioning.pdf.
Requirements:
Microsoft Automatic Graph Layout
http://research.microsoft.com/en-us/projects/msagl/
Gurobi Optimizer
http://www.gurobi.com/
Permutations, Combinations, and...