2-D Pole Balancing with Recurrent Evolutionary Networks

The success of evolutionary methods on standard control learning tasks has created a need for new benchmarks. The classic pole balancing problem is no longer difficult enough to serve as a viable yardstick for measuring the learning efficiency of these systems. In this paper we present a more difficult version to the classic problem where the cart and pole can move in a plane. We demonstrate a neuroevolution system (Enforced Sub-Populations, or ESP) that can solve this difficult problem without velocity information. 1 Introduction The pole-balancing or inverted pendulum problem has long been established as a standard benchmark for artificial learning systems. For over 30 years researchers in fields ranging from control engineering to reinforcement learning have tested their systems on this task [1--3]. There are two primary reasons for this longevity: (1) Pole balancing has intuitive appeal. It is a real-world task that is easy to understand and visualize. It can be performed manually..