Probabilistic Models of the Brain: Perception and Neural Function

By Rajesh P. N. Rao, Barbara Olshausen, Michael S. Lewicki. 2002

418 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/a10ee57d0a61e38cb66617a6a1fbeb4c558fd670

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Deep Learning in Spiking Neural Networks (2018, methodology): “9], [82], [83], [84]. 2) Probabilistic Characterization of Unsupervised STDP: Many studies have provided evidence that at least an approximate Bayesian analysis of sensory stimuli occurs in the brain [85], [86], [87], [88]. In Bayesian inference, hidden causes (such as presence of an object of a par”
Blind Image Quality Assessment Using Joint Statistics of Gradient Magnitude and Laplacian Features (2014, methodology): “This is also true for other bandpass feature extraction methods such as wavelet transform [38], DCT [15], MSCN [14], etc.”
Simple Substrates for Complex Cognition (2008, background): “Again arguing by analogy, unsupervised learning methods applied to the natural statistics of sensory inputs build models that use basis representations to split up complex perceptual scenes into their separate pieces (Rao et al., 2002).”
Off-line replay maintains declarative memories in a model of hippocampal-neocortical interactions (2004, background): “Distributed neocortical representations are acquired and continually adjusted through an unsupervised learning process that alters synaptic connections within and between neocortical areas to reflect the (ever-changing) statistical structure of the inpu”
Exploration from Generalization Mediated by Multiple Controllers (2013, background): “…representations that act as a coordinate system for this low dimensional space so that all natural inputs can be described in terms of these underlying structures (e.g., visual objects and lighting) rather than the full input (i.e., all pixel values) (Hinton and Ghahramani 1997; Rao et al. 2002).”