Abstract
We show how a multi-resolution network can model the development of acuity and coarse-to-fine processing in the mammalian visual cortex. The network adapts to input statistics in an unsupervised manner, and learns a coarse-to-fine representation by using cumulative inhibition of nodes within a network layer. We show that a system of such layers can represent input by hierarchically composing larger parts from smaller components. It can also model aspects of top-down processes, such as image regeneration.
Highlights
In a dynamical environment, it is essential for animals to react quickly, but they need to finely discriminate between stimuli, like those associated with food and mates
We address some criticisms against sparse coding, and suggest some ways they might be resolved in the context of cumulative inhibition
We present a selection of past approaches to multi-resolution representation, and theories on how multiresolution relates to perception
Summary
It is essential for animals to react quickly, but they need to finely discriminate between stimuli, like those associated with food and mates. To accommodate these dual requirements, perceptual networks in mammals represent information in both a coarse and a finely detailed manner. We present the reverse hierarchy theory of multi-resolved representation in particular. Possible mechanisms for such representations are anti-Hebbian learning and sparse coding, which is presented next. A brief overview of the mammalian canonical microcircuit is given These circuits are candidate sites for mediating multi-resolved representation in mammalian brains
Sign-in/Register to view highlights & summary 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.
