Sparse coding via thresholding and local competition in neural circuits (bibtex)

by C.J. Rozell, D.H Johnson, R.G. Baraniuk and B.A. Olshausen

Abstract:

While evidence indicates that neural systems may be employing sparse approximations to represent sensed stimuli, the mechanisms underlying this ability are not understood. We describe a local ly competitive algorithm (LCA) that solves a collection of sparse coding principles minimizing a weighted combination of mean-squared error (MSE) and a coefficient cost function. LCAs are designed to be implemented in a dynamical system composed of many neuron-like elements operating in parallel. These algorithms use thresholding functions to induce local (usually one-way) inhibitory competitions between nodes to produce sparse representations. LCAs produce coefficients with sparsity levels comparable to the most popular centralized sparse coding algorithms while being readily suited for neural implementation. Addi- tionally, LCA coefficients for video sequences demonstrate inertial properties that are both qualitatively and quantitatively more regular (i.e., smoother and more predictable) than the coefficients produced by greedy algorithms.

Reference:

Sparse coding via thresholding and local competition in neural circuitsC.J. Rozell, D.H Johnson, R.G. Baraniuk and B.A. Olshausen. Neural Computation, 20(10), pp. 2526–2563, October 2008.

Bibtex Entry:

@Article{rozell.06c, author = {Rozell, C.J. and Johnson, D.H and Baraniuk, R.G. and Olshausen, B.A.}, title = {Sparse coding via thresholding and local competition in neural circuits}, journal = {Neural Computation}, year = {2008}, volume = {20}, number = {10}, pages = {2526--2563}, month = {October}, abstract = {While evidence indicates that neural systems may be employing sparse approximations to represent sensed stimuli, the mechanisms underlying this ability are not understood. We describe a local ly competitive algorithm (LCA) that solves a collection of sparse coding principles minimizing a weighted combination of mean-squared error (MSE) and a coefficient cost function. LCAs are designed to be implemented in a dynamical system composed of many neuron-like elements operating in parallel. These algorithms use thresholding functions to induce local (usually one-way) inhibitory competitions between nodes to produce sparse representations. LCAs produce coefficients with sparsity levels comparable to the most popular centralized sparse coding algorithms while being readily suited for neural implementation. Addi- tionally, LCA coefficients for video sequences demonstrate inertial properties that are both qualitatively and quantitatively more regular (i.e., smoother and more predictable) than the coefficients produced by greedy algorithms. }, url = {http://siplab.gatech.edu/pubs/rozellNeuralComp2008.pdf} }

Powered by bibtexbrowser