Broad Learning from Narrow Training: A Case Study in Robotic Soccer



next up previous
Next: Introduction

Abstract:

The range of unseen instances that can be successfully classified by a learning algorithm is determined not only by the distribution of the training data, but also by the parameters of the function to be learned. With the right parameters, learning in a certain region of the state space can generalize to completely different areas with no retraining. We demonstrate the power of using well-chosen inputs to (and outputs from) neural networks by conducting experiments in our robotic soccer domain. We train an agent to shoot a moving ball into a goal in a specific situation and end up with a general shooting behavior that is much more widely applicable.

Keywords:
Robotic Soccer, Neural Networks, Flexible Inputs

Tech Report Number:
CMU-CS-95-207


Broad Learning from Narrow Training: A Case Study in Robotic Soccer

Peter Stone and Manuela Veloso





Peter Stone
Wed Nov 8 14:49:26 EST 1995