IIIT Hyderabad Publications
Optimal Spherical Seperability: Towards Optimal Kernel Design
Authors: Garimella Ramamurthy
Report no: IIIT/TR/2016/20
Abstract—In this research paper, the concept of hyperspherical/hyper-ellipsoidal separability is introduced. Method of arriving at the optimal hypersphere (maximizing margin) separating two classes is discussed. By projecting the quantized patterns into higher dimensional space (as in encoders of error correcting code), the patterns are made hyper-spherically separable.Single/multiple layers of spherical/ellipsoidal neurons are proposed for multi-class classification. An associative memory based on hyper-ellipsoidal neuron is proposed. The problem of optimal kernel design is highlighted.
Full report: pdf
Centre for Security, Theory and Algorithms
Copyright © 2009 - IIIT Hyderabad. All Rights Reserved.