Learning Mixture Models of Spatial Coherence

By Suzanna Becker, Geoffrey E. Hinton. 1993

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

How to Represent Part-Whole Hierarchies in a Neural Network (2021, background): “Similarly, early work on learning spatially coherent properties of images used the presence of boundaries to select which expert interpolator to use [Becker and Hinton, 1993].”
Slow feature analysis yields a rich repertoire of complex cell properties. (2005, background): “Ideas and learning rules related to temporal slowness can also be found in the works by Becker and Hinton (1993), O’Reilly and Johnson (1994), and Peng, Sha, Gan, and Wei (1998).”
Unsupervised Feature Learning and Deep Learning: A Review and New Perspectives (2012, background): “The principle of identifying slowly moving/changing factors in temporal/spatial data has been investigated by many (Becker and Hinton, 1993; Wiskott and Sejnowski, 2002; Hurri and Hyv¨arinen, 2003; K¨ording et al. , 2004; Cadieu and Olshausen, 2009) as a principle for ﬁnding useful representations.”
How Auto-Encoders Could Provide Credit Assignment in Deep Networks via Target Propagation (2014, background): “…these principles coincide to some extent (if the forward-going neuron and the backward-going neuron are the same) with the temporal coherence prior (Becker and Hinton, 1993; Wiskott and Sejnowski, 2002; Hurri and Hyv¨arinen, 2003; K ¨ ording et al. , 2004; Cadieu and Olshausen, 2009), sometimes…”
Mutual information maximization: models of cortical self-organization. (1996, methodology): “In joint work with Hinton [ 30 ], when we applied this idea to random dot stereograms, each output unit learned to become tuned to a different range of disparities.”
Representation Learning: A Review and New Perspectives (2012, background): “ deep convolutional networks and wavelets. 8.3 Temporal coherence and slow features The principle of identifying slowly moving/changing factors in temporal/spatial data has been investigated by many (Becker and Hinton, 1993; Wiskott and Sejnowski, 2002; Hurri and Hyv¨arinen, 2003; K ording¨ et al., ”