|
Abstract : |
We describe a form of nonlinear decomposition that is well-suited for efficient encoding of natural signals. Signals are initially decomposed using a bank of linear filters. Each filter response is then rectified and divided by a weighted sum of rectified responses of neighboring filters. We show that this decomposition, with parameters optimized for the statistics of a generic ensemble of natural images or sounds, provides a surprisingly good characterization of the nonlinear response properties of typical neurons in primary visual cortex or auditory nerve, respectively. These results suggest that nonlinear response properties of sensory neurons are not an accident of biological implementation, but serve an important functional role. Signals arising in the natural world are highly structured. To an observer with knowledge of these structures the signals are redundant, because one spatial or temporal portion of a given signal may be predicted from others. Indeed, this is why modern communication technologies are able to efficiently compress and transmit signals. It is widely assumed that neurons in sensory areas of the brain are adapted, through processes of evolution, |
