2015/12/31 by SueYeon Chung, Daniel D. Lee, Haim Sompolinsky
Biochemistry, Genetics and Molecular Biology · Computer Science · Mathematics · Neuroscience · Physics and Astronomy · Psychology · #Algorithm #Artificial intelligence #Artificial neural network #Combinatorics #Computer science #Computer vision #Curse of dimensionality #Decoding methods #Geometry #Invariant (physics) #Mathematics #Neural Networks and Applications #Neural dynamics and brain function #Neuroscience #Object (grammar) #Orientation (vector space) #Pattern recognition (psychology) #Perceptron #Psychology #Representation (politics) #Sensitivity (control systems) #Sensory system #Topology (electrical circuits) #Visual perception and processing mechanisms #cond-mat.dis-nn #cond-mat.stat-mech #cs.NE #q-bio.NC #stat.ML
paper · pdf · doi:10.1103/physreve.93.060301
published as Phys. Rev. E 93, 060301 (R) (2016) · 5 pages, 3 figures, accepted in Physical Review E as Rapid Communication on 14th May. 2016
openalex publication_date 2016/06/30 · arxiv created 2016/08/21 · arxiv updated 2016/08/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/08/05
Objects are represented in sensory systems by continuous manifolds due to sensitivity of neuronal responses to changes in physical features such as location, orientation, and intensity. What makes certain sensory representations better suited for invariant decoding of objects by downstream networks? We present a theory that characterizes the ability of a linear readout network, the perceptron, to classify objects from variable neural responses. We show how the readout perceptron capacity depends on the dimensionality, size, and shape of the object manifolds in its input neural representation.