Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

11

NIPS
1998

favoriteEmaildiscussreport

132views Information Technology» more NIPS 1998»

Learning a Continuous Hidden Variable Model for Binary Data

13 years 5 months ago

Learning a Continuous Hidden Variable Model for Binary Data

Download web.mit.edu

A directed generative model for binary data using a small number of hidden continuous units is investigated. A clipping nonlinearity distinguishes the model from conventional principal components analysis. The relationships between the correlations of the underlying continuous Gaussian variables and the binary output variables are utilized to learn the appropriate weights of the network. The advantages of this approach are illustrated on a translationally invariant binary distribution and on handwritten digit images.

Daniel D. Lee, Haim Sompolinsky

Real-time Traffic

Binary Output Variables | Continuous Gaussian Variables | Directed Generative Model | NIPS 1998 | NIPS 2007 |

claim paper

Related Content

» Continuous Sigmoidal Belief Networks Trained using Slice Sampling

» New dseparation identification results for learning continuous latent variable models

» Context Estimation and Learning Control through Latent Variable Extraction From discrete t...

» Ideal Parent Structure Learning for Continuous Variable Networks

» Learning the Dimensionality of Hidden Variables

» Variable resolution Markov modelling of signal data for image compression

» Quantization of Continuous Input Variables for Binary Classification

» Mapping from Speech to Images Using Continuous State Space Models

» Variational Bayes for Continuous Hidden Markov Models and Its Application to Active Learni...

Post Info
More Details (n/a)

Added	01 Nov 2010
Updated	01 Nov 2010
Type	Conference
Year	1998
Where	NIPS
Authors	Daniel D. Lee, Haim Sompolinsky

Comments (0)