Bilinear Sparse Coding for Invariant Vision

By David B. Grimes, Rajesh P. N. Rao. 2005

109 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Cognitive Neuroscience, Signal Processing

How researchers see this paper

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

In Artificial Intelligence

In Cognitive Neuroscience

In Signal Processing

Citations in the corpus

Disentangling Factors of Variation via Generative Entangling (2012, methodology): “One of the earlier eﬀorts in this direction also used higher-order interactions of latent variables, speciﬁcally bilinear (Tenen-baum and Freeman, 2000; Grimes and Rao, 2005) and multilinear (Vasilescu and Terzopoulos, 2005) models.”
Discovering Hidden Factors of Variation in Deep Networks (2014, methodology): “Previous approaches to separating factors of variation in data, such as content vs. style [3] or form vs. motion [4, 5, 6, 7, 8, 9], have relied upon a bilinear model architecture in which the units representing different factors are combined multiplicatively.”
A Survey of Inductive Biases for Factorial Representation-Learning (2016, background): “ W. The weight on a component is determined by the product of the za and zb coecients of the groups it is a part of. Examples of Bilinear Models. A simple example of a bilinear model can be found in [17], which also includes a sparsity constraint on a and b. Sparsity here means that, for each factor”
Learning to Relate Images (2011, methodology): “Like an auto-encoder, this model can be trained with gradient based optimization.”
Unsupervised learning of visual invariance with temporal coherence (2011, background): “We focus on learning a second layer with temporal coherence.”
A Structured Model of Video Reproduces Primary Visual Cortical Organisation (2009, methodology): “The computational power of a class of models similar to the one in this paper has been investigated by Tenenbaum and Freeman [36], and Grimes and Rao [37].”
Gradient-based learning of higher-order image features (2011, methodology): “When using a single hidden layer with linear activations, the model takes the form of a (conditionally trained) bi-linear model [31] [8], [20], where additional “simple cells” can be used to pre-project inputs before computing multiplicative interactions.”
Stereopsis via deep learning (2013, background): “Recently, several bi-linear (and non-linear) methods to learning image relations have been proposed (for example, [18], [16], [24], [28], [6]).”
Learning to Represent Spatial Transformations with Factored Higher-Order Boltzmann Machines (2010, methodology): “Imposing sparsity constraints on hidden variables has been a common way of obtaining localized filters in static image coding models (for example, Olshausen and Field, 1996) and recently also in bi-linear models of transformations (for example, Grimes and Rao, 2005).”