Next:
Introduction
Up:
Master's Thesis
Previous:
Master's Thesis
Contents
Introduction
Sparse coding
Sparse coding in the cortex
Previous algorithms for sparse coding
Reconstruction error minimisation
Mutual predictability minimisation
Derivation of the new algorithm
Basis of the algorithm
Competition mechanism
Competition between two neurons
Combining the competitions
Finding the outputs
Finding the set of winners
Learning rule
Summary of the algorithm
Extensions of the algorithm
Evaluation of the algorithm
Fixed point iteration of winning strengths
Computational complexity
Simulations
The outputs of the network
Learning artificial data
Learning natural data
Discussion
References
Derivations
Derivation of the equation
Derivation of the equation
About this document ...
Harri Lappalainen
Thu May 9 14:06:29 DST 1996