Studying error propagation on application data structure and hardware

Ozturk, Zuhal; Topcuoglu, Haluk Rahmi; Kandemir, Mahmut

doi:10.1007/s11227-022-04625-x

Cited by 4 publications

(2 citation statements)

References 63 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In statistics, error propagation for high-dimensional spatiotemporal data regression [67,68] and generalization capabilities of the surrogate models for operational optimization [69][70][71] are popular research directions on data interpretability. It is being extensively noticed in artificial intelligence and smart manufacturing domains, such as deep networks and reinforcement learning.…”

Section: Discussionmentioning

confidence: 99%

Data analytics and optimization for smart industry

Tang

Meng

2020

Front. Eng. Manag.

View full text Add to dashboard Cite

Hit rate is a key performance metric in predicting process product quality in integrated industrial processes. It represents the percentage of products accepted by downstream processes within a controlled range of quality. However, optimizing hit rate is a non-convex and challenging problem.To address this issue, we propose a data-driven quasi-convex approach that combines factorial hidden Markov models, multitask elastic net, and quasi-convex optimization. Our approach converts the original non-convex problem into a set of convex feasible problems, achieving an optimal hit rate. We verify the convex optimization property and quasi-convex frontier through Monte Carlo simulations and real-world experiments in steel production. Results demonstrate that our approach outperforms classical models, improving hit rates by at least 41.11% and 31.01% on two real datasets. Furthermore, the quasi-convex frontier provides a reference explanation and visualization for the deterioration of solutions obtained by conventional models.

show abstract

Section: Discussionmentioning

confidence: 99%

Data analytics and optimization for smart industry

Tang

Meng

2020

Front. Eng. Manag.

View full text Add to dashboard Cite

show abstract

“…This feature is clearly manifested when used as number system, for example, residue number system. It is known [17][18][19] that redundant computer system with residue number system contains (15-20)% more equipment PR V than a computer system in a positional binary number system with the same given bit grid without taking into account the addition of secondary redundancy. As preliminary studies have shown, in order to achieve a given level of fault tolerance of the computer system in residue classes, 50% less volume of equipment is required than for system in a positional binary number system.…”

Section: Problem Statementmentioning

confidence: 99%

The Structure of the Computer System in the Residual Classes

Koshman

Krasnobayev

Nikolsky

et al. 2023

A.I.S.

View full text Add to dashboard Cite

The subject of the article is the formulation and solution of the inverse problem of optimal redundancy in the system of residual classes (RNS) based on the use of the dynamic programming method. The solution of this problem makes it possible to improve the reliability of the operation of computer systems and components (CSC) in the RNS. The purpose of the article is to increase the reliability of the functioning of CSC, which are built on the basis of the use of RNS, without reducing the speed of calculations, as well as to calculate and compare the reliability, in terms of the probability of failure-free operation, of CSC in RNS and a tripled computing system that operates in a positional binary number system (PNS). Tasks: to analyze the influence of the number system used on the reliability of the CSC, taking into account the primary and secondary redundancy; to synthesize a computing system in RNS for a l-byte bit grid based on the use of the passive fault tolerance method (constant structural redundancy); formulate and solve the inverse problem of optimal redundancy in RNS based on the use of the dynamic programming method; to check the correctness of the results obtained, calculate the conditional amount of computer system equipment in the residual classes; evaluate the efficiency of using RNS to improve reliability when building a redundant CSC in relation to a redundant CSC in the PNS. Research methods: methods of analysis and synthesis of computer systems, number theory, coding theory in RNS, reliability theory. The following results are obtained. The paper shows that the use of PNS as a number system does not allow a radical increase in the performance and reliability of CSC. In this regard, the article developed the concept of using RNS as a number system for constructing a CSC. Based on this, the inverse problem of optimal redundancy in RNS is formulated and solved. Conclusions. As shown by the results of calculations and comparative analysis, the use of RNS provides a higher reliability of the CSC than the majority three-channel computing system in the PNS. The obtained research results can be used for the synthesis of fault-tolerant computer structures in RNS.

show abstract

Research on safety verification methods of static data of train control systems based on deep association rules

Wang,

2024

J Supercomput

View full text Add to dashboard Cite

Studying error propagation on application data structure and hardware

Cited by 4 publications

References 63 publications

Data analytics and optimization for smart industry

Data analytics and optimization for smart industry

The Structure of the Computer System in the Residual Classes

Research on safety verification methods of static data of train control systems based on deep association rules

Contact Info

Product

Resources

About