Opposition-Based Learning: A New Scheme for Machine Intelligence

TLDR

Opposition-based learning, a new scheme for machine intelligence, is introduced and is based on estimates and counter-estimates, weights and opposite weights, and actions versus counter-actions. Examples are provided, and preliminary results are presented. Possibilities for extensions of existing learning algorithms are discussed.

Abstract

Opposition-based learning as a new scheme for machine intelligence is introduced. Estimates and counter-estimates, weights and opposite weights, and actions versus counter-actions are the foundation of this new approach. Examples are provided. Possibilities for extensions of existing learning algorithms are discussed. Preliminary results are provided