Geoffrey E. Hinton

TL;DR: A novel Gibbs-Langevin sampling algorithm that outperforms existing methods like Gibbs sampling and a modified contrastive divergence algorithm so that one can generate images with GRBMs starting from noise, enabling direct comparison of GRB Ms with deep generative models.

TL;DR: It is shown that by using neural networks to provide better starting points, the search time can be significantly reduced on a character recognition task.

TL;DR: The human brain is very different from a conventional digital computer; it relies on massive parallelism rather than raw speed and it stores long-term knowledge by modifying the way its processing elements interact rather than by setting bits in a passive, general purpose memory.

TL;DR: The spectral gradient descent (SGD) method as mentioned in this paper uses information contained in the leading eigenvalues of a data affinity matrix to modify the steps taken during a gradient-based optimization procedure and is able to speed up the optimization and to help dimensionality reduction methods find better local minima of their objective functions.