scispace - formally typeset
Search or ask a question
Journal Article

Optimization Methods

About: This article is published in IEEE Transactions on Systems, Man, and Cybernetics.The article was published on 1979-02-01 and is currently open access. It has received 98 citations till now. The article focuses on the topics: Stochastic optimization & Robust optimization.
Citations
More filters
Journal ArticleDOI
TL;DR: The scope covers gradient descent and polynomial line search, from backpropagation through conjugate gradients and quasi Newton methods, and shows a consensus among researchers that adaptive step gains can stabilize and accelerate convergence and that a good starting weight set improves both the training speed and the learning quality.
Abstract: We survey research of recent years on the supervised training of feedforward neural networks. The goal is to expose how the networks work, how to engineer them so they can learn data with less extraneous noise, how to train them efficiently, and how to assure that the training is valid. The scope covers gradient descent and polynomial line search, from backpropagation through conjugate gradients and quasi Newton methods. There is a consensus among researchers that adaptive step gains (learning rates) can stabilize and accelerate convergence and that a good starting weight set improves both the training speed and the learning quality. The training problem includes both the design of a network function and the fitting of the function to a set of input and output data points by computing a set of coefficient weights. The form of the function can be adjusted by adjoining new neurons and pruning existing ones and setting other parameters such as biases and exponential rates. Our exposition reveals several useful results that are readily implementable.

178 citations

Journal ArticleDOI
TL;DR: In this article, a five-level-five-factor central composite (circumscribed) design (CCD) approach-based response surface methodology (RSM) analysis was applied to statistically specify the effect of important process variables, such as initial PVP polymer concentration (6−14 wt %), applied voltage (10−22kV), flow rate (4−16 μlit/min), nozzle-collector distance (10 −18 cm), and molar ratio of boron to hafnium (2.2−5.8)
Abstract: Electrospinning in order to achieve the new nanofibrous composites by using inorganic precursor and in the lack of defects, like beads or bubbles and having diameter uniformity characteristic is challenging. To synthesis of HfB2, repeatability and homogeneity of raw electrospun fibers of polyvinylpyrrolidone-boron-hafnium (PVP-B-Hf) complex from PVP, H3BO3, and HfCl4 precursors are essential to further applications; for example in synthesis of HfB2. A five-level-five-factor central composite (circumscribed) design (CCD) approach-based response surface methodology (RSM) analysis was applied to statistically specify the effect of important process variables, namely initial PVP polymer concentration (6–14 wt %), applied voltage (10–22 kV), flow rate (4–16 μlit/min), nozzle-collector distance (10–18 cm), and molar ratio of boron to hafnium (2.2–5.8) on the key response process output variables such as average diameter, quality, and uniformity of the nanofibrous composites. By using this rotatable design, a total of 50 electrospinning experimental data was fitted. The significance of the factors and their interactions were verified by using the analysis of variance (ANOVA) with 95% of confidence level (p

130 citations

Journal ArticleDOI
TL;DR: In this paper, the authors examine the role of informatics in several important materials subfields, highlighting significant contributions to date and identifying known shortcomings. And they specifically focus on the difference between the correlative approach of classical data science and the causative approach for physical sciences.

103 citations

Journal ArticleDOI
TL;DR: A special method based on the linearization with respect to the basic nonconvexity of a d.c. minimization problem with a simple feasible set is developed and a variant of local search for a general d. c. optimization problem is considered.

49 citations

Journal ArticleDOI
TL;DR: This work presents an approach based on Genetic Algorithms (GA) to optimize image segmentation parameters by using the performance scores from object-based classification, thus allowing to assess the adequacy of a segmented image in relation to the classification problem.

39 citations