Sciweavers

4757 search results - page 240 / 952
» Generalized Posynomial Performance Modeling
Sort
View
JMLR
2012
13 years 9 months ago
Krylov Subspace Descent for Deep Learning
In this paper, we propose a second order optimization method to learn models where both the dimensionality of the parameter space and the number of training samples is high. In ou...
Oriol Vinyals, Daniel Povey
SIGMETRICS
2012
ACM
197views Hardware» more  SIGMETRICS 2012»
13 years 9 months ago
Fluid limit of an asynchronous optical packet switch with shared per link full range wavelength conversion
We consider an asynchronous all optical packet switch (OPS) where each link consists of N wavelength channels and a pool of C ≤ N full range tunable wavelength converters. Under...
Benny Van Houdt, Luca Bortolussi
SIGMETRICS
2008
ACM
161views Hardware» more  SIGMETRICS 2008»
15 years 6 months ago
Bound analysis of closed queueing networks with workload burstiness
Burstiness and temporal dependence in service processes are often found in multi-tier architectures and storage devices and must be captured accurately in capacity planning models...
Giuliano Casale, Ningfang Mi, Evgenia Smirni
SAMOS
2007
Springer
16 years 21 days ago
A Study of Energy Saving in Customizable Processors
Abstract. Embedded systems are special purpose systems which perform predefined tasks with very specific requirements like high performance, low volume or low power. Most of the ...
Paolo Bonzini, Dilek Harmanci, Laura Pozzi
SDM
2008
SIAM
165views Data Mining» more  SDM 2008»
15 years 8 months ago
On the Dangers of Cross-Validation. An Experimental Evaluation
Cross validation allows models to be tested using the full training set by means of repeated resampling; thus, maximizing the total number of points used for testing and potential...
R. Bharat Rao, Glenn Fung