In this paper we introduce a new underlying probabilistic model for principal component analysis (PCA). Our formulation interprets PCA as a particular Gaussian process prior on a ...
The principle of maximizing mutual information is applied to learning overcomplete and recurrent representations. The underlying model consists of a network of input units driving...
In thispaper, we present a novelhybridarchitecture forcontinuousspeech recognition systems. It consists of a continuous HMM system extended by an arbitrary neural network that is ...
Embedding algorithms search for low dimensional structure in complex data, but most algorithms only handle objects of a single type for which pairwise distances are specified. Thi...
Amir Globerson, Gal Chechik, Fernando C. Pereira, ...
An auditory "scene", composed of overlapping acoustic sources, can be viewed as a complex object whose constituent parts are the individual sources. Pitch is known to be...