Design of Trustworthy Cyber-Physical-Social Systems with Discrete Bayesian Optimization

2021 ◽  
pp. 1-31
Author(s):  
Yan Wang
Keyword(s):  
Network Optimization ◽  
Network Architecture ◽  
Optimization Problem ◽  
Social Systems ◽  
Systems Design ◽  
Optimization Method ◽  
Bayesian Optimization ◽  
Security And Privacy ◽  
Quantitative Metrics ◽  
Share Information

Abstract Cyber-physical-social systems (CPSS) with highly integrated functions of sensing, actuation, computation, and communication are becoming the mainstream consumer and commercial products. The performance of CPSS heavily relies on the information sharing between devices. Given the extensive data collection and sharing, security and privacy are of major concerns. Thus one major challenge of designing those CPSS is how to incorporate the perception of trust in product and systems design. Recently a trust quantification method was proposed to measure trustworthiness of CPSS by quantitative metrics of ability, benevolence, and integrity. The CPSS network architecture can be optimized by choosing a subnet such that the trust metrics are maximized. The combinatorial network optimization problem however is computationally challenging. Most of the available global optimization algorithms for solving such problems are heuristic methods. In this paper, a surrogate-based discrete Bayesian optimization method is developed to perform network design, where the most trustworthy CPSS network with respect to a reference node is formed to collaborate and share information with. The applications of ability and benevolence metrics in design optimization of CPSS architecture are demonstrated.

Design of Trustworthy Cyber-Physical-Social Systems With Discrete Bayesian Optimization

2020 ◽  
Author(s):  
Yan Wang
Keyword(s):  
Data Collection ◽  
Information Sharing ◽  
Social Systems ◽  
Systems Design ◽  
Optimization Method ◽  
Bayesian Optimization ◽  
Security And Privacy ◽  
Quantification Method ◽  
Quantitative Metrics ◽  
Share Information

Abstract Cyber-physical-social systems (CPSS) with highly integrated functions of sensing, actuation, computation, and communication are becoming the mainstream consumer and commercial products. The performance of CPSS heavily relies on the information sharing between devices. Given the extensive data collection and sharing, security and privacy are of major concerns. Thus one major challenge of designing those CPSS is how to incorporate the perception of trust in product and systems design. Recently a trust quantification method was proposed to measure trustworthiness of CPSS by quantitative metrics of ability, benevolence, and integrity. In this paper, the applications of ability and benevolence metrics in design optimization of CPSS architecture are demonstrated. A Bayesian optimization method is developed to perform trust based CPSS network design, where the most trustworthy network with respect to a reference node can be selected to collaborate and share information with.

scholarly journals Deep Convolutional Neural Network Optimization for Defect Detection in Fabric Inspection

Sensors ◽  
2021 ◽  
Vol 21 (21) ◽  
pp. 7074
Author(s):  
Chao-Ching Ho ◽  
Wei-Chi Chou ◽  
Eugene Su
Keyword(s):  
Image Processing ◽  
Deep Learning ◽  
Embedded System ◽  
Network Model ◽  
Defect Detection ◽  
Network Optimization ◽  
Network Architecture ◽  
Bayesian Optimization ◽  
Original Network ◽  
Network Pruning

This research is aimed to detect defects on the surface of the fabric and deep learning model optimization. Since defect detection cannot effectively solve the fabric with complex background by image processing, this research uses deep learning to identify defects. However, the current network architecture mainly focuses on natural images rather than the defect detection. As a result, the network architecture used for defect detection has more redundant neurons, which reduces the inference speed. In order to solve the above problems, we propose network pruning with the Bayesian optimization algorithm to automatically tune the network pruning parameters, and then retrain the network after pruning. The training and detection process uses the above-mentioned pruning network to predict the defect feature map, and then uses the image processing flow proposed in this research for the final judgment during fabric defect detection. The proposed method is verified in the two self-made datasets and the two public datasets. In the part of the proposed network optimization results, the Intersection over Union (IoU) of four datasets are dropped by 1.26%, 1.13%, 1.21%, and 2.15% compared to the original network model, but the inference time is reduced to 20.84%, 40.52%, 23.02%, and 23.33% of the original network model using Geforce 2080 Ti. Furthermore, the inference time is also reduced to 17.56%, 37.03%, 19.67%, and 22.26% using the embedded system AGX Xavier. After the image processing part, the accuracy of the four datasets can reach 92.75%, 94.87%, 95.6%, and 81.82%, respectively. In this research, Yolov4 is also trained with fabric defects, and the results showed this model are not conducive to detecting long and narrow fabric defects.

The Improvement of Newton Method and the Validation of Optimization Idea of Blind Walking Repeatedly

2011 ◽  
Vol 250-253 ◽  
pp. 4061-4064
Author(s):  
Chun Ling Zhang
Keyword(s):  
Saddle Point ◽  
Newton Method ◽  
Optimization Problem ◽  
Initial Point ◽  
Optimization Method ◽  
Maximum Point ◽  
Optimal Result ◽  
Feasible Domain ◽  
Parallel Arithmetic ◽  
Newton Direction

The existence of maximum point, oddity point and saddle point often leads to computation failure. The optimization idea is based on the reality that the optimum towards the local minimum related the initial point. After getting several optimal results with different initial point, the best result is taken as the final optimal result. The arithmetic improvement of multi-dimension Newton method is improved. The improvement is important for the optimization method with grads convergence rule or searching direction constructed by grads. A computational example with a saddle point, maximum point and oddity point is studied by multi-dimension Newton method, damped Newton method and Newton direction method. The importance of the idea of blind walking repeatedly is testified. Owing to the parallel arithmetic of modernistic optimization method, it does not need to study optimization problem with seriate feasible domain by modernistic optimization method.

Parameter study and optimization design of a hat oblique tunnel portal

2021 ◽  
pp. 095440972110140
Author(s):  
Zijian Guo ◽  
Tanghong Liu ◽  
Wenhui Li ◽  
Yutao Xia
Keyword(s):  
High Speed ◽  
Optimization Design ◽  
Pareto Front ◽  
Optimization Problem ◽  
Optimization Method ◽  
Design Parameters ◽  
Parameter Study ◽  
Buffer Structure ◽  
Tunnel Entrance ◽  
The Ideal

The present work focuses on the aerodynamic problems resulting from a high-speed train (HST) passing through a tunnel. Numerical simulations were employed to obtain the numerical results, and they were verified by a moving-model test. Two responses, [Formula: see text] (coefficient of the peak-to-peak pressure of a single fluctuation) and[Formula: see text] (pressure value of micro-pressure wave), were studied with regard to the three building parameters of the portal-hat buffer structure of the tunnel entrance and exit. The MOPSO (multi-objective particle swarm optimization) method was employed to solve the optimization problem in order to find the minimum [Formula: see text] and[Formula: see text]. Results showed that the effects of the three design parameters on [Formula: see text] were not monotonous, and the influences of[Formula: see text] (the oblique angle of the portal) and [Formula: see text] (the height of the hat structure) were more significant than that of[Formula: see text] (the angle between the vertical line of the portal and the hat). Monotonically decreasing responses were found in [Formula: see text] for [Formula: see text] and[Formula: see text]. The Pareto front of [Formula: see text] and[Formula: see text]was obtained. The ideal single-objective optimums for each response located at the ends of the Pareto front had values of 1.0560 for [Formula: see text] and 101.8 Pa for[Formula: see text].

scholarly journals A New Hyper-Parameter Optimization Method for Power Load Forecast Based on Recurrent Neural Networks

Algorithms ◽  
2021 ◽  
Vol 14 (6) ◽  
pp. 163
Author(s):  
Yaru Li ◽  
Yulai Zhang ◽  
Yongping Cai
Keyword(s):  
Neural Networks ◽  
Parameter Optimization ◽  
Recurrent Neural Networks ◽  
Critical Role ◽  
Optimization Method ◽  
Bayesian Optimization ◽  
Power Load ◽  
Gradient Calculation ◽  
Truncated Newton ◽  
Selection Of

The selection of the hyper-parameters plays a critical role in the task of prediction based on the recurrent neural networks (RNN). Traditionally, the hyper-parameters of the machine learning models are selected by simulations as well as human experiences. In recent years, multiple algorithms based on Bayesian optimization (BO) are developed to determine the optimal values of the hyper-parameters. In most of these methods, gradients are required to be calculated. In this work, the particle swarm optimization (PSO) is used under the BO framework to develop a new method for hyper-parameter optimization. The proposed algorithm (BO-PSO) is free of gradient calculation and the particles can be optimized in parallel naturally. So the computational complexity can be effectively reduced which means better hyper-parameters can be obtained under the same amount of calculation. Experiments are done on real world power load data,where the proposed method outperforms the existing state-of-the-art algorithms,BO with limit-BFGS-bound (BO-L-BFGS-B) and BO with truncated-newton (BO-TNC),in terms of the prediction accuracy. The errors of the prediction result in different models show that BO-PSO is an effective hyper-parameter optimization method.

scholarly journals An Optimization Method for the Configuration of Inter Array Cables for Floating Offshore Wind Farm

2017 ◽  
Author(s):  
Yann Poirette ◽  
Martin Guiton ◽  
Guillaume Huwart ◽  
Delphine Sinoquet ◽  
Jean Marc Leroy
Keyword(s):  
Wind Farm ◽  
Optimization Problem ◽  
Initial Point ◽  
Trust Region ◽  
Optimization Method ◽  
Global Optimum ◽  
Offshore Wind ◽  
Finite Element Software ◽  
Constrained Problems ◽  
Floating Offshore Wind Turbines

IFP Energies nouvelles (IFPEN) is involved for many years in various projects for the development of floating offshore wind turbines. The commercial deployment of such technologies is planned for 2020. The present paper proposes a methodology for the numerical optimization of the inter array cable configuration. To illustrate the potential of such an optimization, results are presented for a case study with a specific floating foundation concept [1]. The optimization study performed aims to define the least expensive configuration satisfying mechanical constraints under extreme environmental conditions. The parameters to be optimized are the total length, the armoring, the stiffener geometry and the buoyancy modules. The insulated electrical conductors and overall sheath are not concerned by this optimization. The simulations are carried out using DeepLines™, a Finite Element software dedicated to simulate offshore floating structures in their marine environment. The optimization problem is solved using an IFPEN in-house tool, which integrates a state of the art derivative-free trust region optimization method extended to nonlinear constrained problems. The latter functionality is essential for this type of optimization problem where nonlinear constraints are introduced such as maximum tension, no compression, maximum curvature and elongation, and the aero-hydrodynamic simulation solver does not provide any gradient information. The optimization tool is able to find various local feasible extrema thanks to a multi-start approach, which leads to several solutions of the cable configuration. The sensitivity to the choice of the initial point is demonstrated, illustrating the complexity of the feasible domain and the resulting difficulty in finding the global optimum configuration.

Trajectory Optimization Applied to Space Rendezvous

2013 ◽  
Vol 756-759 ◽  
pp. 3466-3470
Author(s):  
Xu Min Song ◽  
Qi Lin
Keyword(s):  
Simulated Annealing ◽  
Optimization Model ◽  
Trajectory Optimization ◽  
Optimization Problem ◽  
Optimization Method ◽  
Nonlinear Constraints ◽  
Design Variables ◽  
Adaptive Simulated Annealing ◽  
Simulation Results

The trajcetory plan problem of spece reandezvous mission was studied in this paper using nolinear optimization method. The optimization model was built based on the Hills equations. And by analysis property of the design variables, a transform was put forward , which eliminated the equation and nonlinear constraints as well as decreaseing the problem dimensions. The optimization problem was solved using Adaptive Simulated Annealing (ASA) method, and the rendezvous trajectory was designed.The method was validated by simulation results.

Stochastic Bayesian optimization for predicting borderline knock

2021 ◽  
pp. 146808742110652
Author(s):  
Jian Tang ◽  
Anuj Pal ◽  
Wen Dai ◽  
Chad Archer ◽  
James Yi ◽  
...  
Keyword(s):  
Fuel Economy ◽  
Learning Algorithm ◽  
Feedforward Control ◽  
Model Development ◽  
Kriging Model ◽  
Optimization Method ◽  
Surrogate Models ◽  
Bayesian Optimization ◽  
Control Parameters ◽  
Operation Conditions

Engine knock is an undesirable combustion that could damage the engine mechanically. On the other hand, it is often desired to operate the engine close to its borderline knock limit to optimize combustion efficiency. Traditionally, borderline knock limit is detected by sweeping tests of related control parameters for the worst knock, which is expensive and time consuming, and also, the detected borderline knock limit is often used as a feedforward control without considering its stochastic characteristics without compensating current engine operational condition and type of fuel used. In this paper, stochastic Bayesian optimization method is used to obtain a tradeoff between stochastic knock intensity and fuel economy. The log-nominal distribution of knock intensity signal is converted to Gaussian one using a proposed map to satisfy the assumption for Kriging model development. Both deterministic and stochastic Kriging surrogate models are developed based on test data using the Bayesian iterative optimization process. This study focuses on optimizing two competing objectives, knock intensity and indicated specific fuel consumption using two control parameters: spark and intake valve timings. Test results at two different operation conditions show that the proposed learning algorithm not only reduces required time and cost for predicting knock borderline but also provides control parameters, based on trained surrogate models and the corresponding Pareto front, with the best fuel economy possible.

scholarly journals Wireless network optimization method based on cognitive cycle using machine learning

2018 ◽  
Vol 7 (7) ◽  
pp. 278-283 ◽  
Author(s):  
Koji Oshima ◽  
Takumu Kobayashi ◽  
Yuki Taenaka ◽  
Kaori Kuroda ◽  
Mikio Hasegawa
Keyword(s):  
Machine Learning ◽  
Wireless Network ◽  
Network Optimization ◽  
Optimization Method ◽  
Wireless Network Optimization
Sign in / Sign up

Export Citation Format

Share Document