Restricted Boltzmann machines were developed using binary stochastic hidden units. These can be generalized by replacing each binary unit by an infinite number of copies that all ...
The goal of the visual correspondence problem is to establish a connectivity pattern (a mapping) between two images such that features projected from the same scene point are conn...
We study a class of functional which can be used for matching objects which can be represented as mappings from a fixed interval, I, to some "feature space." This class o...
In this paper, an unsupervised learning algorithm, neighborhood linear embedding (NLE), is proposed to discover the intrinsic structures such as neighborhood relationships, global ...
Shuzhi Sam Ge, Feng Guan, Yaozhang Pan, Ai Poh Loh
This paper describes a new method for building compact context-dependency transducers for finite-state transducer-based ASR decoders. Instead of the conventional phonetic decision...