Recurrent neural networks serve as black-box models for nonlinear dynamical systems identification and time series prediction. Training of recurrent networks typically minimizes t...
Faceted classification allows one to model applications with complex classification hierarchies using orthogonal dimensions. Recent work has examined the use of faceted classi...
We recently proposed a new approach to parallelization, by decomposing the time domain, instead of the conventional space domain. This improves latency tolerance, and we demonstrat...
This paper presents a global optimization operator for arbitrary meshes. The global optimization operator is composed of two main terms, one part is the global Laplacian operator ...
Ligang Liu, Chiew-Lan Tai, Zhongping Ji, Guojin Wa...
Recently the LATL has undertaken the development of a multilingual translation system based on a symbolic parsing technology and on a transfer-based translation model. A crucial c...