Abstract
Barlow's redundancy reduction hypothesis is applied using techniques developed from signal processing theory to derive the one-dimensional ganglion and simple cell kernels. The resulting closed-form expression for the ganglion cell kernel reduces redundancy over the entire range of signal-to-noise ratios, and resembles the phenomenological kernel. Significantly, it exhibits observed nontrivial dependence of the cell's parameters on background luminosity. The one-dimensional simple cell kernel is deduced by requiring that it maintain the redundancy reduction achieved by the ganglion cells despite the presence of intrinsic noise introduced through transmission along the optic nerve Neural update algorithms which converge to these cell profiles, starting from any initial set of synaptic strengths, are also derived. These learning algorithms turn out to be anti-Hebbian.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
