Adding reinforcement learning features to the neural-gas method | IEEE Conference Publication | IEEE Xplore