scholarly journals BaKer-Nets: Bayesian Random Kernel Mapping Networks

2020 ◽  
Author(s):  
Hui Xue ◽  
Zheng-Fan Wu
Keyword(s):  
Long Range ◽  
Bayesian Learning ◽  
Learning Processes ◽  
The Novel ◽  
Local Minima ◽  
Feature Spaces ◽  
Core Components ◽  
The Cost ◽  
Spectral Kernel ◽  
Kernel Mapping

Recently, deep spectral kernel networks (DSKNs) have attracted wide attention. They consist of periodic computational elements that can be activated across the whole feature spaces. In theory, DSKNs have the potential to reveal input-dependent and long-range characteristics, and thus are expected to perform more competitive than prevailing networks. But in practice, they are still unable to achieve the desired effects. The structural superiority of DSKNs comes at the cost of the difficult optimization. The periodicity of computational elements leads to many poor and dense local minima in loss landscapes. DSKNs are more likely stuck in these local minima, and perform worse than expected. Hence, in this paper, we propose the novel Bayesian random Kernel mapping Networks (BaKer-Nets) with preferable learning processes by escaping randomly from most local minima. Specifically, BaKer-Nets consist of two core components: 1) a prior-posterior bridge is derived to enable the uncertainty of computational elements reasonably; 2) a Bayesian learning paradigm is presented to optimize the prior-posterior bridge efficiently. With the well-tuned uncertainty, BaKer-Nets can not only explore more potential solutions to avoid local minima, but also exploit these ensemble solutions to strengthen their robustness. Systematical experiments demonstrate the significance of BaKer-Nets in improving learning processes on the premise of preserving the structural superiority.

scholarly journals Learning Deeper Non-Monotonic Networks by Softly Transferring Solution Space

2021 ◽  
Author(s):  
Zheng-Fan Wu ◽  
Hui Xue ◽  
Weimin Bai
Keyword(s):  
High Frequency ◽  
Bayesian Learning ◽  
Solution Space ◽  
Fourier Space ◽  
Local Minima ◽  
Error Feedback ◽  
Learning Framework ◽  
Empirical Risk ◽  
Core Components ◽  
In The Beginning

Different from popular neural networks using quasiconvex activations, non-monotonic networks activated by periodic nonlinearities have emerged as a more competitive paradigm, offering revolutionary benefits: 1) compactly characterizing high-frequency patterns; 2) precisely representing high-order derivatives. Nevertheless, they are also well-known for being hard to train, due to easily over-fitting dissonant noise and only allowing for tiny architectures (shallower than 5 layers). The fundamental bottleneck is that the periodicity leads to many poor and dense local minima in solution space. The direction and norm of gradient oscillate continually during error backpropagation. Thus non-monotonic networks are prematurely stuck in these local minima, and leave out effective error feedback. To alleviate the optimization dilemma, in this paper, we propose a non-trivial soft transfer approach. It smooths their solution space close to that of monotonic ones in the beginning, and then improve their representational properties by transferring the solutions from the neural space of monotonic neurons to the Fourier space of non-monotonic neurons as the training continues. The soft transfer consists of two core components: 1) a rectified concrete gate is constructed to characterize the state of each neuron; 2) a variational Bayesian learning framework is proposed to dynamically balance the empirical risk and the intensity of transfer. We provide comprehensive empirical evidence showing that the soft transfer not only reduces the risk of non-monotonic networks on over-fitting noise, but also helps them scale to much deeper architectures (more than 100 layers) achieving the new state-of-the-art performance.

Insights into Transdermal Drug Delivery: Approaches for Redressal of a Burgeoning Issue of Osteoporosis

2020 ◽  
Vol 20 (10) ◽  
pp. 1682-1695
Author(s):  
Foziyah Zakir ◽  
Kanchan Kohli ◽  
Farhan J. Ahmad ◽  
Zeenat Iqbal ◽  
Adil Ahmad
Keyword(s):  
Drug Delivery ◽  
Transdermal Delivery ◽  
Oral Delivery ◽  
Calcium Absorption ◽  
Early Stage ◽  
Osteoporotic Fractures ◽  
The Novel ◽  
Systemic Delivery ◽  
Primary Focus ◽  
The Cost

Osteoporosis is a progressive bone disease that remains unnoticed until a fracture occurs. It is more predominant in the older age population, particularly in females due to reduced estrogen levels and ultimately limited calcium absorption. The cost burden of treating osteoporotic fractures is too high, therefore, primary focus should be treatment at an early stage. Most of the marketed drugs are available as oral delivery dosage forms. The complications, as well as patient non-compliance, limit the use of oral therapy for prolonged drug delivery. Transdermal delivery systems seem to be a promising approach for the delivery of anti-osteoporotic active moieties. One of the confronting barriers is the passage of drugs through the SC layers followed by penetration to deeper dermal layers. The review focuses on how anti-osteoporotic drugs can be molded through different approaches so that they can be exploited for the skin to systemic delivery. Insights into the various challenges in transdermal delivery and how the novel delivery system can be used to overcome these have also been detailed.

scholarly journals The Effect of Heat Treatment on the Microstructure and Mechanical Properties of the Novel Low-Cost Ti-3Al-5Mo-4Cr-2Zr-1Fe Alloy

Materials ◽  
2020 ◽  
Vol 13 (17) ◽  
pp. 3798
Author(s):  
Meng Sun ◽  
Dong Li ◽  
Yanhua Guo ◽  
Ying Wang ◽  
Yuecheng Dong ◽  
...  
Keyword(s):  
Mechanical Properties ◽  
Heat Treatment ◽  
Aging Time ◽  
Volume Fraction ◽  
Low Cost ◽  
Solution Treatment ◽  
Solution Temperature ◽  
The Novel ◽  
Α Phase ◽  
The Cost

In order to reduce the cost of titanium alloys, a novel low-cost Ti-3Al-5Mo-4Cr-2Zr-1Fe (Ti-35421) titanium alloy was developed. The influence of heat treatment on the microstructure characteristics and mechanical properties of the new alloy was investigated. The results showed that the microstructure of Ti-35421 alloy consists of a lamina primary α phase and a β phase after the solution treatment at the α + β region. After aging treatment, the secondary α phase precipitates in the β matrix. The precipitation of the secondary α phase is closely related to heat treatment parameters—the volume fraction and size of the secondary α phase increase when increasing the solution temperature or aging time. At the same solution temperature and aging time, the secondary α phase became coarser, and the fraction decreased with increasing aging temperature. When Ti-35421 alloy was solution-treated at the α + β region for 1 h with aging surpassing 8 h, the tensile strength, yield strength, elongation and reduction of the area were achieved in a range of 1172.7–1459.0 MPa, 1135.1–1355.5 MPa, 5.2–11.8%, and 7.5–32.5%, respectively. The novel low-cost Ti-35421 alloy maintains mechanical properties and reduces the cost of materials compared with Ti-3Al-5Mo-5V-4Cr-2Zr (Ti-B19) alloy.

scholarly journals On the Use of LoRa Technology for Logic Selectivity in MV Distribution Networks

Energies ◽  
2018 ◽  
Vol 11 (11) ◽  
pp. 3079 ◽  
Author(s):  
Leopoldo Angrisani ◽  
Francesco Bonavolontà ◽  
Annalisa Liccardo ◽  
Rosario Schiano Lo Moriello
Keyword(s):  
Cost Effectiveness ◽  
Long Range ◽  
Distribution Networks ◽  
Fault Isolation ◽  
Optimized Design ◽  
Medium Voltage ◽  
Different Types ◽  
The Times ◽  
The Cost ◽  
Types Of Faults

In this paper, a logic selectivity system based on Long Range (LoRa) technology for the protection of medium-voltage (MV) networks is proposed. The development of relays that communicate with each other using LoRa allows for the combination of the cost-effectiveness and ease of installation of wireless networks with long-range coverage and reliability. The realized demonstrator to assess the proposed system is also presented in the paper; based on different types of faults and different locations, the times needed for clearing a fault and restoring the network were estimated from repeated experiments. The obtained results confirm that, with an optimized design of transmitted packets and of protocol characteristics, LoRa communication grants fault management that meets the criteria of logic selectivity, with fault isolation occurring within the maximum allowed time.

scholarly journals Application Layer Packet Processing Using PISA Switches

Sensors ◽  
2021 ◽  
Vol 21 (23) ◽  
pp. 8010
Author(s):  
Ismail Butun ◽  
Yusuf Tuncel ◽  
Kasim Oztoprak
Keyword(s):  
Streaming Data ◽  
The Novel ◽  
Application Layer ◽  
Domain Specific ◽  
Switch Architecture ◽  
Novel Approach ◽  
Data Processor ◽  
Application Identification ◽  
The Cost ◽  
Telecommunication Operators

This paper investigates and proposes a solution for Protocol Independent Switch Architecture (PISA) to process application layer data, enabling the inspection of application content. PISA is a novel approach in networking where the switch does not run any embedded binary code but rather an interpreted code written in a domain-specific language. The main motivation behind this approach is that telecommunication operators do not want to be locked in by a vendor for any type of networking equipment, develop their own networking code in a hardware environment that is not governed by a single equipment manufacturer. This approach also eases the modeling of equipment in a simulation environment as all of the components of a hardware switch run the same compatible code in a software modeled switch. The novel techniques in this paper exploit the main functions of a programmable switch and combine the streaming data processor to create the desired effect from a telecommunication operator perspective to lower the costs and govern the network in a comprehensive manner. The results indicate that the proposed solution using PISA switches enables application visibility in an outstanding performance. This ability helps the operators to remove a fundamental gap between flexibility and scalability by making the best use of limited compute resources in application identification and the response to them. The experimental study indicates that, without any optimization, the proposed solution increases the performance of application identification systems 5.5 to 47.0 times. This study promises that DPI, NGFW (Next-Generation Firewall), and such application layer systems which have quite high costs per unit traffic volume and could not scale to a Tbps level, can be combined with PISA to overcome the cost and scalability issues.

scholarly journals The γTuRC Revisited: A Comparative Analysis of Interphase and Mitotic Human γTuRC Redefines the Set of Core Components and Identifies the Novel Subunit GCP8

2010 ◽  
Vol 21 (22) ◽  
pp. 3963-3972 ◽  
Author(s):  
Neus Teixidó-Travesa ◽  
Judit Villén ◽  
Cristina Lacasa ◽  
Maria Teresa Bertran ◽  
Marco Archinti ◽  
...  
Keyword(s):  
Mass Spectrometry ◽  
Comparative Analysis ◽  
Protein Complex ◽  
The Novel ◽  
Microtubule Network ◽  
Microtubule Polymerization ◽  
Open Ring ◽  
Core Components ◽  
Ring Complexes ◽  
Ring Shaped Structure

The γ-tubulin complex is a multi-subunit protein complex that nucleates microtubule polymerization. γ-Tubulin complexes are present in all eukaryotes, but size and subunit composition vary. In Drosophila, Xenopus, and humans large γ-tubulin ring complexes (γTuRCs) have been described, which have a characteristic open ring-shaped structure and are composed of a similar set of subunits, named γ-tubulin, GCPs 2-6, and GCP-WD in humans. Despite the identification of these proteins, γTuRC function and regulation remain poorly understood. Here we establish a new method for the purification of native human γTuRC. Using mass spectrometry of whole protein mixtures we compared the composition of γTuRCs from nonsynchronized and mitotic human cells. Based on our analysis we can define core subunits as well as more transient interactors such as the augmin complex, which associates specifically with mitotic γTuRCs. We also identified GCP8/MOZART2 as a novel core subunit that is present in both interphase and mitotic γTuRCs. GCP8 depletion does not affect γTuRC assembly but interferes with γTuRC recruitment and microtubule nucleation at interphase centrosomes without disrupting general centrosome structure. GCP8-depleted cells do not display any obvious mitotic defects, suggesting that GCP8 specifically affects the organization of the interphase microtubule network.

Modeling of Long-Range Reverberation Signal for Rough Bottom Consisting of Polygon Facets

2018 ◽  
Vol 26 (04) ◽  
pp. 1850041
Author(s):  
Youngmin Choo ◽  
Woojae Seong
Keyword(s):  
Long Range ◽  
Order Approximation ◽  
Higher Order ◽  
Ocean Bottom ◽  
Analytic Surface ◽  
Surface Integral ◽  
Computational Burden ◽  
Analytic Integration ◽  
The Cost ◽  
Delay Difference

To acquire a stable reverberation signal from an irregular ocean bottom, we derive the analytic surface integral of a scattered signal using Stokes’ theorem while approximating the bottom using a combination of polygon facets. In this approach, the delay difference in the elemental scattering area is considered, while the representative delay is used for the elemental scattering area in the standard reverberation model. Two different reverberation models are applied to a randomly generated rough bottom, which is composed of triangular facets. Their results are compared, and the scheme using analytic integration shows a converged reverberation signal, even with a large elemental scattering area, at the cost of an additional computational burden caused by a higher order approximation in the surface integral of the scattered signals.

Application of Dynamic Programming in Long-Range Pipeline Facility Planning and in Assessing Off-Design and Off-Flow-Forecast Trajectories

2008 ◽  
Author(s):  
K. K. Botros ◽  
C. Foy ◽  
B. Chmilar
Keyword(s):  
Dynamic Programming ◽  
Long Range ◽  
Supply And Demand ◽  
Planning Horizon ◽  
Design Flow ◽  
Facility Planning ◽  
The Cost ◽  
The Impact ◽  
Planning Problems ◽  
Flow Forecast

Dynamic programming (DP) inherently provides a methodology for evaluating a series of decisions in order to determine an optimal policy or path forward. The methodology basically enumerates and evaluates alternative states over the planning horizon in formulating the optimum strategy. In the present work, the concept of DP has been applied to pipeline long-range facility planning problems, and further extended to allow evaluation of nth optimum pipeline facility deployments based on cost and/or probabilities of constraints. The best four options were further analyzed considering uncertainties in the cost elements and the resulting economic risk associated with each optimum path. This paper presents the theory behind the extension of the DP methodology to pipeline long-range facility-planning problems over a planning horizon that considers inherent uncertainties in gas supply and demand as well as a range of available facility options. Uncertainties in the size and location of the required facilities to handle the forecast volumes, and associated variances in their respective cost to build and operate the various facilities, are all accounted for. The problem is further complicated by the possible changes in the expected flow from that forecast during design and the resulting penalties associated with the under- or over-sizing of facilities. It was demonstrated that it is important that the off-design flow forecast be evaluated to determine the impact of future variability or changes. The value that the organization can derive from being able to quantify the benefit (or penalty) of forecast uncertainty and over- or under-building long-range facilities, is significant.

A Critical Review of German Long-Range Rocket Development

1946 ◽  
Vol 50 (427) ◽  
pp. 483-525 ◽  
Author(s):  
W. G. A. Perring
Keyword(s):  
Research And Development ◽  
Long Range ◽  
Critical Review ◽  
Set Up ◽  
The Cost

Serious rocket development was started in Germany in the years 1929–1930 by a few groups of private inventors. This work attracted the attention of the Army Weapons Group in 1933, and in 1937–1938 a special research and development station was set up at Peenemunde at the cost of 300 million marks, and the work was transferred from Berlin to this new station at about that time.

Sign in / Sign up

Export Citation Format

Share Document