A Modified Conjugate Gradient Approach for Reliability-Based Design Optimization

Wang, Zhiming; Zhang, Yafei; Song, Yalong

doi:10.1109/access.2020.2966661

Cited by 8 publications

(5 citation statements)

References 32 publications

(29 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In a subsequent work, Frady et al proposed a method for gradient-based learning in HDC that utilized iterative projections and local linearizations to facilitate learning in high-dimensional spaces [34]. Building upon these developments, Wang et al presented a gradient-based HDC algorithm for clustering and classification tasks, which employed a convex optimization formulation to enhance HDC's performance in these applications [35]. Moreover, Su et al developed a gradient-based HDC algorithm for deep learning, illustrating the potential of gradient-based methods in improving the robustness and expressiveness of HDC models [36].…”

Section: B Gradient Based Hdc Methodsmentioning

confidence: 99%

“…For example, Frady and Somer's work involves iterative projections and local linearizations which can be expensive. Similarly Wang's convex optimization formulation for clustering and classification tasks can result in increased computational overhead [35] Prior works have also explored the use Holographic Reduced Representations (HRR), a family models for gradient based learning tasks. A notable attempt to capitalize on the symbolic properties of HRR was made by Nickel et al [38], who utilized binding operations to link elements within a knowledge graph.…”

Section: B Gradient Based Hdc Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Label Classification With Hyperdimensional Representations

Chandrasekaran,

Asgareinjad,

Morris

et al. 2023

IEEE Access

View full text Add to dashboard Cite

Hyperdimensional computing (HDC) is a computational paradigm that leverages the mathematical properties of high-dimensional vector spaces to manipulate data as symbolic entities using a set of neurally plausible operations. Although HDC has demonstrated remarkable success in cognitive tasks, its potential in complex applications such as multi-label classificati has yet to be explored. In this research paper, we introduce three approaches to multi-label classification that strike a balance between computational efficiency and accuracy, based on the complexity of the problem. The first approach we propose is Power Set HD, a transformation method that is ideal for small-scale multi-label classification with label cardinality less than four and label set size less than ten. The second approach, One-vs-All HD, is another transformation method that is suitable for slightly more complex tasks with higher label cardinality, providing a better efficiency-accuracy trade-off over Power Set HD. However, due to the expensive linear complexity scaling of One-vs-All HD, we propose a novel neural approach called TinyXML HD for extreme scale tasks. This method learns hyperdimensional representations by decomposing the learning problem into multiple sub-problems, which are solved neurally through gradient-based optimization. Importantly, TinyXML HD fixes the output size of the model to the dimensionality of the hypervector, regardless of the label size, thereby scaling only by a small constant when evaluated on datasets with extremely large label spaces. Our approaches offer a valuable trade-off between computational efficiency and accuracy. We show that our methods provide a speedup of 16-60x on state of the art datasets, while maintaining comparable accuracy. Furthermore, our methods yield models that are 56x smaller on medium-scale tasks and up to 836x smaller on extreme-scale datasets, which is a significant reduction in model size while still achieving high accuracy.

show abstract

Section: B Gradient Based Hdc Methodsmentioning

confidence: 99%

Section: B Gradient Based Hdc Methodsmentioning

confidence: 99%

Multi-Label Classification With Hyperdimensional Representations

Chandrasekaran,

Asgareinjad,

Morris

et al. 2023

IEEE Access

View full text Add to dashboard Cite

show abstract

“…RBDO may utilize a number of different techniques: a gradient approach to increase the effectiveness of RBDO in extreme nonlinearities [38], sequential optimization to solve for quadratic models [39], [40], and single loop approaches to assist in long-term time-based optimization problems [41], [42]. The single loop approach focuses on approximating the reliability constraint of a system to estimate how long the model will remain efficient or run without failure [43].…”

Section: List Of Figuresmentioning

confidence: 99%

Design optimization of a wind tunnel force balance using stepwise response surface method

Chaisson

Lee²,

Burns³

2022

J Mech Sci Technol

View full text Add to dashboard Cite

applied to a body during a wind tunnel test. The balance is equipped with strain gauges that measure structural deformations for evaluation of the forces and moments being applied to the body. The force balance is required to have structural safety under a large roll moment, due to design changes with the testing apparatus. The balance must be optimized to allow for this increased roll moment while guaranteeing the balance will not fail and the gauges within the balance will generate an acceptable gauge reading.

show abstract

“…With the development of mechanical systems towards high precision and reliability, it is essential to consider uncertainties in engineering design. Reliability-based design optimization (RBDO) is one of the most powerful design methods considering uncertainties, which seeks an optimum design that satisfies probability constraints [1]. Uncertainties can be generally classified into aleatory and epistemic uncertainties [2].…”

Section: Introductionmentioning

confidence: 99%

Kriging-Based Performance Measure Function Approximation Method for Hybrid Reliability-Based Design Optimization

et al. 2023

View full text Add to dashboard Cite

Hybrid reliability-based design optimization (HRBDO) can provide an effective way to obtain the optimum design in the presence of both random and interval variables. HRBDO is typically described as a nested optimization model. It is computationally expensive when directly solving the HRBDO problem by the nested optimization method. To address this issue, this paper develops an efficient decoupled HRBDO method that aims at performance measure function approximation by the adaptive Kriging model (KPMFA). The proposed KPMFA method includes three main blocks, namely hybrid inverse reliability analysis, performance measure function approximation, and equivalent deterministic optimization. In KPMFA, the adaptation of the adaptive chaos control (ACC) algorithm for inverse reliability analysis that accommodates interval variables is developed. Moreover, an adaptive strategy with two-stage of enrichment for the Kriging model is developed to approximate performance measure functions on the region of interest. Then, the optimization can be proceeded using the Kriging model of performance measure functions. Finally, five illustrative HRBDO problems are investigated to demonstrate the accuracy and efficiency of the proposed KPMFA method.INDEX TERMS Hybrid reliability-based design optimization (HRBDO); random and interval variables; adaptive kriging model; performance measure function approximation

show abstract

A Modified Conjugate Gradient Approach for Reliability-Based Design Optimization

Cited by 8 publications

References 32 publications

Multi-Label Classification With Hyperdimensional Representations

Multi-Label Classification With Hyperdimensional Representations

Design optimization of a wind tunnel force balance using stepwise response surface method

Kriging-Based Performance Measure Function Approximation Method for Hybrid Reliability-Based Design Optimization

Contact Info

Product

Resources

About