Predicting System Performance by Interpolation Using a High-Dimensional Delaunay Triangulation

Prediction of Stock Trend Based on Deep Belief Networks

Applied Mechanics and Materials ◽

10.4028/www.scientific.net/amm.644-650.5538 ◽

2014 ◽

Vol 644-650 ◽

pp. 5538-5541 ◽

Cited By ~ 1

Author(s):

Zhao Yang Xu ◽

Li Na Tang ◽

Chun Peng Tian

Keyword(s):

Stock Market ◽

System Performance ◽

Stock Price ◽

High Dimensional ◽

Belief Networks ◽

Deep Belief Networks ◽

Softmax Regression ◽

The Past ◽

Market Experiments ◽

Novel Model

The deep belief networks (DBNs) are introduced to predict the stock trend. By stacking three RBMs and a softmax regression, a novel model of the stock price is developed to extract the high dimensional feature and predict the trend of the stock market. Experiments on Yahoo stock market of the past three years are implemented, and the performance of the multilayer perceptron (MLP) and DBN with different input dimensions are compared. A detailed discussion is given on the improvement of the system performance.

Download Full-text

Sensitivity Derivatives for Flexible Sensorimotor Learning

Neural Computation ◽

10.1162/neco.2008.04-07-507 ◽

2008 ◽

Vol 20 (8) ◽

pp. 2085-2111 ◽

Cited By ~ 18

Author(s):

M. N. Abdelghani ◽

T. P. Lillicrap ◽

D. B. Tweed

Keyword(s):

System Performance ◽

Adaptive Controller ◽

Sensorimotor Control ◽

Neural Systems ◽

High Dimensional ◽

Sensorimotor Learning ◽

Work Spaces ◽

Sensitivity Derivatives ◽

The Brain ◽

Information Transport

To learn effectively, an adaptive controller needs to know its sensitivity derivatives—the variables that quantify how system performance depends on the commands from the controller. In the case of biological sensorimotor control, no one has explained how those derivatives themselves might be learned, and some authors suggest they are not learned at all but are known innately. Here we show that this knowledge cannot be solely innate, given the adaptive flexibility of neural systems. And we show how it could be learned using forms of information transport that are available in the brain. The mechanism, which we call implicit supervision, helps explain the flexibility and speed of sensorimotor learning and our ability to cope with high-dimensional work spaces and tools.

Download Full-text

Expected Length of the Voronoi Path in a High Dimensional Poisson–Delaunay Triangulation

Discrete & Computational Geometry ◽

10.1007/s00454-017-9866-y ◽

2017 ◽

Vol 60 (1) ◽

pp. 200-219

Author(s):

Pedro Machado Manhães de Castro ◽

Olivier Devillers

Keyword(s):

Delaunay Triangulation ◽

High Dimensional ◽

Expected Length

Download Full-text

The Propagation of Vortex Beams in Random Mediums

10.5772/intechopen.101061 ◽

2021 ◽

Author(s):

Sekip Dalgac ◽

Kholoud Elmabruk

Keyword(s):

Angular Momentum ◽

Orbital Angular Momentum ◽

Optical Communications ◽

System Performance ◽

Topological Charge ◽

Dark Spot ◽

High Dimensional ◽

Free Space Optical ◽

Phase Singularity ◽

Vortex Beams

Vortex beams acquire increasing attention due to their unique properties. These beams have an annular spatial profile with a dark spot at the center, the so-called phase singularity. This singularity defines the helical phase structure which is related to the topological charge value. Topological charge value allows vortex beams to carry orbital angular momentum. The existence of orbital angular momentum offers a large capacity and high dimensional information processing which make vortex beams very attractive for free-space optical communications. Besides that, these beams are well capable of reducing turbulence-induced scintillation which leads to better system performance. This chapter introduces the research conducted up to date either theoretically or experimentally regarding vortex beam irradiance, scintillation, and other properties while propagating in turbulent mediums.

Download Full-text