Investigating Neural Activation Effects on Deep Belief Echo-State Networks for Prediction Toward Smart Ocean Environment Monitoring

Author(s):  
Zhigang Li ◽  
Jialin Wang ◽  
Difei Cao ◽  
Yingqi Li ◽  
Xiaochuan Sun ◽  
...  
2013 ◽  
Vol 67 (1) ◽  
pp. 177-189 ◽  
Author(s):  
Zhi Zhao ◽  
Kefeng Ji ◽  
Xiangwei Xing ◽  
Huanxin Zou ◽  
Shilin Zhou

Ship surveillance is important for maritime security and safety. It plays important roles in many applications including ocean environment monitoring, search and rescue, anti-piracy and military reconnaissance. Among various sensors used for maritime surveillance, space-borne Synthetic Aperture Radar (SAR) is valued for its high resolution over wide swaths and all-weather working capabilities. However, the state-of-the-art algorithms for ship detection and identification do not always achieve a satisfactory performance. With the rapid development of space-borne Automatic Identification System (AIS), near real-time and global surveillance has become feasible. However, not all ships are equipped with or operate AIS. Space-borne SAR and AIS are considered to be complementary, and ship surveillance using an integrated combination has attracted much attention. In order to summarize the achievements and present references for further research, this paper attempts to explicitly review the developments in previous research as the basis of a brief introduction to space-borne SAR and AIS.


2017 ◽  
Vol 28 (07) ◽  
pp. 1750095 ◽  
Author(s):  
M. Andrecut

Reservoir Computing (RC) refers to a Recurrent Neural Network (RNNs) framework, frequently used for sequence learning and time series prediction. The RC system consists of a random fixed-weight RNN (the input-hidden reservoir layer) and a classifier (the hidden-output readout layer). Here, we focus on the sequence learning problem, and we explore a different approach to RC. More specifically, we remove the nonlinear neural activation function, and we consider an orthogonal reservoir acting on normalized states on the unit hypersphere. Surprisingly, our numerical results show that the system’s memory capacity exceeds the dimensionality of the reservoir, which is the upper bound for the typical RC approach based on Echo State Networks (ESNs). We also show how the proposed system can be applied to symmetric cryptography problems, and we include a numerical implementation.


2010 ◽  
Vol 43 (20) ◽  
pp. 277-282 ◽  
Author(s):  
Yuji Tomizawa ◽  
Masayoshi Toda

2019 ◽  
Vol 348 ◽  
pp. 158-168 ◽  
Author(s):  
Shengke Wang ◽  
Lu Liu ◽  
Liang Qu ◽  
Changyin Yu ◽  
Yujuan Sun ◽  
...  

2012 ◽  
Author(s):  
R. Montirosso ◽  
S. Moriconi ◽  
B. Riccardi ◽  
G. Reni ◽  
F. Arrigoni ◽  
...  

2014 ◽  
pp. 99-122
Author(s):  
M. Levin ◽  
K. Matrosova

The paper considers monitoring of environmental change as the central element of environmental regulation. Monitoring, as each kind of principalagent relations, easily gives rise to corruptive behavior. In the paper we analyze economic models of environmental monitoring with high costs, incomplete information and corruption. These models should be the elements of environmental economics and are needed to create an effective system of nature protection measures.


Sign in / Sign up

Export Citation Format

Share Document