Sparse coding of sensory inputs.

By Barbara Olshausen, David J. Field. 2004

1,546 citations recorded by the source corpus.

Subfields: Cellular and Molecular Neuroscience, Cognitive Neuroscience, Representation & Self-Supervised Learning, Neuroscience

How researchers see this paper

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

In Cellular and Molecular Neuroscience

In Cognitive Neuroscience

In Representation & Self-Supervised Learning

In Neuroscience

Citations in the corpus

Feature Purification: How Adversarial Training Performs Robust Deep Learning (2020, methodology): “We use a data model called sparse coding, which is a popular model to model image, text and speech data [11, 12, 69, 77, 78, 100, 105, 106].”
Learning to Linearize Under Uncertainty (2015, background): “For instance, sparse coding implicitly assumes a locally linear model of the data manifold [9].”
How does the brain solve visual object recognition? (2012, neutral): “There are a number of promising candidate ideas and algorithmic classes to consider (e.g., Hinton and Salakhutdinov, 2006; Olshausen and Field, 2004; Wiskott and Sejnowski, 2002).”
Toward a unified theory of efficient, predictive, and sparse coding (2017, background): “A hallmark of natural stimuli is their sparse latent structure (18, 22, 25, 26): stimulus fragments can be constructed from a set of primitive features (e.”
Sparse Coding via Thresholding and Local Competition in Neural Circuits (2008, background): “These populations are typically very overcomplete (Olshausen and Field, 2004), allowing great ﬂexibility in the representation of a stimulus.”
Coupling between neuronal firing rate, gamma LFP, and BOLD fMRI is related to interneuronal correlations. (2007, background): “, excitatory neuronal activity in the hippocampus [43]), gamma LFP and BOLD responses will appear to be less correlated to the dynamics of spiking activity in individual cells.”
Efficient coding of natural sounds (2002, background): “In recent years, studies that applied experimental, computational and theoretical methods [100] could show that one major mechanism involved in the neurosensory information processing is “sparse coding”.”
Feed-Forward versus Feedback Inhibition in a Basic Olfactory Circuit (2015, background): “Sparse codes support learning and memory by reducing interference between memory traces, increase storage capacity, and save energy expended by action potentials [16, 18, 50, 51].”
Network Structure within the Cerebellar Input Layer Enables Lossless Sparse Encoding (2014, background): “This increases the separation between the patterns (Olshausen and Field, 2004), thereby enabling downstream cerebellar circuits to perform more effective associative learning (Albus, 1971; D’Angelo and De Zeeuw, 2009; Marr, 1969; Medina and Mauk, 2000; Schweighofer et al., 2001; Tyrrell and…”
Why Neurons Have Thousands of Synapses, a Theory of Sequence Memory in Neocortex (2015, background): “However, robust recognition is possible if the patterns to be recognized are sparse; i.e. few neurons are active relative to the population (Olshausen and Field, 2004).”
Population coding under normalization (2009, background): “…dependencies of neural activity, and have put forward the idea that normalization may serve to optimize the representation of natural signals (Fairhall, Lewen, Bialek, & de Ruyter Van Steveninck, 2001; Olshausen & Field, 1996a, 1996b, 2004; Ruderman & Bialek, 1994; Schwartz & Simoncelli, 2001).”
Odor representations in olfactory cortex: "sparse" coding, global inhibition, and oscillations. (2009, background): “…is proposed to promote the efficient coding of sensory information in the brain by having a relatively small fraction of neurons within a large population active at any given time (Hromadka et al., 2008; Laurent , 2002; Olshausen and Field, 2004; Rolls and Tovee, 1995; Willmore and Tolhurst, 2001).”
Dendrites of dentate gyrus granule cells contribute to pattern separation by controlling sparsity (2016, background): “Experimental studies have shown that only a small population of GCs, 5%, are active in a single context (Marrone et al., 2011; Satvat et al., 2011; Danielson et al., 2016), a phenomenon termed sparse coding (Olshausen and Field, 2004).”
What Learning Systems do Intelligent Agents Need? Complementary Learning Systems Theory Updated. (2016, neutral): “These are endpoints on a continuum quanti ﬁ ed by a measure called sparsity, where ‘ population ’ sparsity indexes the proportion of neurons that ﬁ re in response to a given stimulus/location, and ‘ lifetime ’ sparsity indexes the proportion of stimuli to which a single neuron responds [26,153,155].”