It is often useful for a robot to construct a spatial representation of its environment from experiments and observations, in other words, to learn a map of its environment by exp...
Thomas Dean, Dana Angluin, Kenneth Basye, Sean P. ...
A new learning algorithmis derived which performs online stochastic gradient ascent in the mutual informationbetween outputs and inputs of a network. In the absence of a priori kn...
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...
We propose a unified neural network architecture and learning algorithm that can be applied to various natural language processing tasks including: part-of-speech tagging, chunki...
We introduce a novel approach to modeling the dynamics of human facial motion induced by the action of speech for the purpose of synthesis. We represent the trajectories of a numbe...