A copy of this work was available on the public web and has been preserved in the Wayback Machine. The capture dates from 2017; you can also visit the original URL.
The file type is application/pdf
.
Filters
A normalized gradient descent algorithm for nonlinear adaptive filters using a gradient adaptive step size
2001
IEEE Signal Processing Letters
A fully adaptive normalized nonlinear gradient descent (FANNGD) algorithm for online adaptation of nonlinear neural filters is proposed. ...
For rigor, the remainder of the truncated Taylor series expansion within the expression for the adaptive learning rate is made adaptive and is updated using gradient descent. ...
A Normalized Gradient Descent Algorithm for Nonlinear Adaptive Filters Using a Gradient Adaptive Step Size Danilo P. Mandic, Andrew I. ...
doi:10.1109/97.969448
fatcat:blpwcsls55gwlfb3wp4srlxczq
Gradient-Sensitive Optimization for Convolutional Neural Networks
2021
Computational Intelligence and Neuroscience
Convolutional neural networks (CNNs) are effective models for image classification and recognition. Gradient descent optimization (GD) is the basic algorithm for CNN model optimization. ...
Our algorithm is a supplement to the existing gradient descent algorithms, which can be combined with many other existing gradient descent algorithms to improve the efficiency of iteration, speed up the ...
Preliminaries e basic algorithm for CNN model optimization is GD. ...
doi:10.1155/2021/6671830
fatcat:xegos3yn25fyxlpnaa2tbnt64q
An Improved CMA-ES for Solving Large Scale Optimization Problem
[chapter]
2020
Lecture Notes in Computer Science
Comparative experiments have been done on state-of-the-art algorithms. The results proved the effectiveness and efficiency of GI-ES for large scale optimization problems. ...
To solve this problem, this paper proposes an improved CMA-ES, called GI-ES, for large-scale optimization problems. ...
Effectiveness of the Gradient Information. The adaptation of the mutation strength is crucial for evolutionary calculation. ...
doi:10.1007/978-3-030-53956-6_34
fatcat:psbhdxnmbvazrd5myb6fwa5ipq
Hyperspherical parametrization for unit-norm based adaptive IIR filtering
1999
IEEE Signal Processing Letters
We propose a hyperspherical parameterization to convert the unit-norm-constrained optimization into an unconstrained optimization. ...
The bias problem associated with equation error based adaptive infinite impulse response (IIR) filtering can be surmounted by imposing a unit-norm constraint on the autoregressive (AR) coefficients. ...
INTRODUCTION T RADITIONALLY, finite impulse response (FIR) structures have been used for adaptive filters, due to their simplicity. ...
doi:10.1109/97.803434
fatcat:pb5lx6igxjgjfmp6wgpxq7qvzy
A global least mean square algorithm for adaptive IIR filtering
1998
IEEE transactions on circuits and systems - 2, Analog and digital signal processing
Combining this approximation of the gradient with the LMS algorithm results in a stochastic global optimization algorithm for adaptive IIR filtering. ...
Derivation of Gradient Estimate The key to implementing any algorithm for adaptive filtering (7) is the development of an on-line gradient estimate r(n; ): Here we propose to apply the SAS derived single-sided ...
The algorithm and simulation results of the compact neural-network-based CDMA receiver are described in this brief. ...
doi:10.1109/82.664244
fatcat:l6w5f6tblva4hl5benddfzs75a
An adaptive optimal strategy based on the combination of the dynamic-Q optimization method and response surface methodology
2005
IEEE transactions on magnetics
The dynamic-Q optimization method is combined with an interpolating moving least-squares approximation-based response surface model to design an efficient adaptive strategy for solving computationally ...
The proposed optimal strategy is validated by comparing its performances in finding the solutions of other common optimal methods on two different kinds of problems. ...
For the sake of completeness, a brief introduction about IMLS is given in the following paragraphs. ...
doi:10.1109/tmag.2005.846031
fatcat:7lxfcblqrne3tldndp5ajhp5yy
Efficient Full-Matrix Adaptive Regularization
[article]
2020
arXiv
pre-print
We also provide a novel theoretical analysis for adaptive regularization in non-convex optimization settings. ...
The core of our algorithm, termed GGT, consists of the efficient computation of the inverse square root of a low-rank matrix. ...
Acknowledgments We are grateful to Yoram Singer, Tomer Koren, Nadav Cohen, and Sanjeev Arora for helpful discussions. ...
arXiv:1806.02958v2
fatcat:vyzeqvt7bbedrn2tyfdjhsat5a
Page 377 of Automation and Remote Control Vol. 34, Issue 3
[page]
1973
Automation and Remote Control
Introduction
At present, there is no lack of various algorithms for finding the unconditional extremum of a functional J(c) which defines an optimality criterion. ...
ADAPTIVE SYSTEMS
PSEUDOGRADIENT ADAPTATION AND TRAINING ALGORITHMS
B. T. Polyak and Ya. Z. ...
An Improved Adam Optimization Algorithm Combining Adaptive Coefficients and Composite Gradients Based on Randomized Block Coordinate Descent
2023
Computational Intelligence and Neuroscience
An improved Adam optimization algorithm combining adaptive coefficients and composite gradients based on randomized block coordinate descent is proposed to address issues of the Adam algorithm such as ...
Simulation experiments on two standard datasets for classification show that the convergence speed and accuracy of the proposed algorithm are higher than those of the six gradient descent methods, and ...
Acknowledgments Tis work was supported by the National Natural Science Foundation of China (Grant nos. 42002138 and 62172352), Natural Science Foundation of Heilongjiang Province (Grant no. ...
doi:10.1155/2023/4765891
pmid:36660559
pmcid:PMC9845049
fatcat:hrjclg6msbhmtpnvlomkuwc6gi
Comparison of Adaptive Ant Colony Optimization for Image Edge Detection of Leaves Bone Structure
2018
Emitter: International Journal of Engineering Technology
In this research, Adaptive Ant Colony Optimization algorithm is proposed for edge image detection of leaf bone structure. ...
that allows for an edge based on the value of the image gradient. ...
INTRODUCTION Plants are the most substansial item of life on earth. The plant is useful as a supplier of oxygen for breathing, as foodstuff, fuel, medicine, cosmetics and more. ...
doi:10.24003/emitter.v6i2.306
fatcat:5yhh2y2e2rgj5bmbrwswbbagfy
Adaptive high order stochastic descent algorithms
2022
Zenodo
After a brief introduction to this framework, we introduce in this talk a new approach, called SGD-G2, which is a high order Runge-Kutta stochastic descent algorithm; the procedure allows for step adaptation ...
One of the most known among them, the Stochastic Gradient Descent (SGD), has been extended in various ways resulting in Adam, Nesterov, momentum, etc. ...
Figure : : Figure: Left: Numerical results (over the first 5 epochs) for the SGD-G2 algorithm on the FMNIST dataset with several choices of the initial learning rate h 0 ; right: SGD , SGD-G2 and Adam ...
doi:10.5281/zenodo.7257153
fatcat:zoapff5eubdntkq3ectebgjaoa
An improved algorithm for radar adaptive beamforming based on machine learning
2019
Journal of Physics, Conference Series
In order to improve the performance of adaptive beamforming, this paper firstly reviews the classical LMS algorithm and then the machine learning optimization algorithm. ...
The Least Mean Square Algorithm (LMS) is a simple and easy algorithm for adaptive digital beamforming. ...
Introduction Adaptive digital beamforming is an important branch of digital signal processing. ...
doi:10.1088/1742-6596/1325/1/012114
fatcat:ubjx3h3fbfbjpgejf6ahfitjae
Enhancing Performance of a Deep Neural Network: A Comparative Analysis of Optimization Algorithms
2020
Advances in Distributed Computing and Artificial Intelligence Journal
Adopting the most suitable optimization algorithm (optimizer) for a Neural Network Model is among the most important ventures in Deep Learning and all classes of Neural Networks. ...
In this paper, we will experiment with seven of the most popular optimization algorithms namely: sgd, rmsprop, adagrad, adadelta, adam, adamax and nadam on four unrelated datasets discretely, to conclude ...
Adaptive Gradient Algorithm (Adagrad) Adaptive Gradient Algorithm (Adagrad) is very similar to stochastic gradient descent algorithm but unlikely uses adaptive gradients to improve robustness as shown ...
doi:10.14201/adcaij2020927990
fatcat:mo7gwxkcujf5fadwwiwoef3xpq
A Modified Bio Inspired
2018
International Journal of Applied Metaheuristic Computing
(CPSO) algorithm, self adaptive penalty function genetic algorithm (SAPFGA) and mutable smart bee algorithm (MSBA), for optimal design of truss structures with dynamic frequency constraints. ...
Therefore, an algorithm for automation of constraint shape and size design of truss structures is proposed here. ...
INTRODUCTION Generally, two different paradigms are taken into account for optimizing engineering systems such as truss design problems. ...
doi:10.4018/ijamc.2018010105
fatcat:jcyynwmmrjde5c2s7aivixnbbu
Improved Binary Forward Exploration: Learning Rate Scheduling Method for Stochastic Optimization
[article]
2022
arXiv
pre-print
and the most successful adaptive learning rate algorithm e.g. ...
The Adaptive version of BFE has also been discussed thereafter. ...
Improved BFE of gradient change Algorithm 3 Improved BFE of gradient change, the proposed algorithm in non-adaptive learning rate automation for stochastic optimization. ...
arXiv:2207.04198v3
fatcat:zkwvjcegevdnrkcyfvn5oxsx5m
« Previous
Showing results 1 — 15 out of 368,935 results