What Is the Goal of Sensory Coding?

By David J. Field. 1994

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 neural networks: An overview (2014, neutral): “Many UL methods are designed to maximize entropy-related, information-theoretic (Boltzmann, 1909; Kullback & Leibler, 1951; Shannon, 1948) objectives (e.g., Amari, Cichocki, & Yang, 1996; Barlowet al., 1989;Dayan&Zemel, 1995;Deco&Parra, 1997; Field, 1994; Hinton, Dayan, Frey, & Neal, 1995; Linsker, ”
Constrained Nonnegative Matrix Factorization for Hyperspectral Unmixing (2009, background): “Thus, sparse coding [26], which refers to a representational scheme where only a few units are effectively used to represent an input vector, can be adopted to deal with the abundancesparseness problem.”
Iterative kernel principal component analysis for image modeling (2005, methodology): “In contrast to linear PCA, KPCA is capable of capturing part of the higher-order statistics which are particularly important for encoding image structure [11].”
Blind deconvolution in scientific imaging & computational photography (2011, background): “(2006) combined the variational approach of Miskin and MacKay (2000) with natural image statistics (Field, 1994).”
Visual perception and the statistical properties of natural scenes. (2008, methodology): “Variants of this efficient coding hypothesis have been widely proposed and evaluated (Atick & Redlich 1992; Attneave 1954; Barlow 1961, 2001; Field 1994; Laughlin 1981; van Hateren 1992).”
Removing Camera Shake from a Single Photograph (2009, background): “…although images of real-world scenes vary greatly in their absolute color distributions, they obey heavytailed distributions in their gradients [Field 1994]: the distribution of gradients has most of its mass on small values but gives significantly more probability to large values than a…”
Unsupervised learning of granule cell sparse codes enhances cerebellar adaptive control. (2001, background): “For supervised learning in a subsequent neural layer, sparse codes combine the advantages of both local and distributed codes without their drawbacks.(22) In a local code, only one cell is active at any time (see Fig.”
Image Deblurring and Super-Resolution by Adaptive Sparse Domain Selection and Adaptive Regularization (2010, background): “, edges, line segments and other elementary features), and these primitives are qualitatively similar in form to simple cell receptive fields [61]–[63].”
Dappled photography: mask enhanced cameras for heterodyned light fields and coded aperture refocusing (2007, background): “It has been shown that real-world images obey heavy tail distributions in their gradients [Field 1994].”
Dictionary Learning Algorithms for Sparse Representation (2003, methodology): “Can the belief that supergaussian priors, P p (x), are appropriate for finding sparse solutions to equation 1.3 (Field, 1994; Olshausen & Field, 1996) be clarified or made rigorous?”
Generalization in Interactive Networks: The Benefits of Inhibitory Competition and Hebbian Learning (2001, background): “Another way of thinking about the beneﬁts of inhibitory competition comes from the idea that, given the general structure of the natural environment, sparse distributed representations (i.e., with relatively few units active at a time) are particularly useful (Barlow, 1989; Field, 1994).”
Dictionaries for Sparse Representation Modeling (2010, background): “Indeed, this concept of compaction will later be replaced with sparsity, though the two are closely related [3].”
Blind deconvolution using a normalized sparsity measure (2011, unknown): “But since images are highly non-Gaussian [5], many approaches [6, 13, 14] instead use `p-norms on the gradients, with 0.”