Sciweavers

NIPS
1996
13 years 5 months ago
Continuous Sigmoidal Belief Networks Trained using Slice Sampling
Real-valued random hidden variables can be useful for modelling latent structure that explains correlations among observed variables. I propose a simple unit that adds zero-mean G...
Brendan J. Frey
IDA
2009
Springer
13 years 11 months ago
Image Denoising Using Sparse Representations
The problem of removing white zero-mean Gaussian noise from an image is an interesting inverse problem to be investigated in this paper through sparse and redundant representations...
Seyyed Majid Valiollahzadeh, Hamed Firouzi, Massou...