Modify Training Directions in Function Space to Reduce Generalization Error
CoRR, 2023 We propose theoretical analyses of a modified natural gradient descent method in the neural network function space based on the eigendecompositions of neural tangent kernel and Fisher information matrix.
Yu, Yi, Chen, Boyu, Lu, Wenlian
core +2 more sources
modified error function with added terms for the backpropagation algorithm
, 2004 Dalian Univ Technol, Chinese Univ Hong Kong, IEEE Circuits & Syst Soc, IEEE Computat Intelligence Soc, IEEE Control Syst Soc, Robot & Automat SocWe have noted that the local minima problem in the back-propagation algorithm is usually caused ...
Tang Z, Bi WX, Zong ZL, Wang XG
core +2 more sources
The error bounds of Gauss quadrature formulae for the modified weight functions of Chebyshev type [PDF]
Applied Mathematics and Computation, 2020 In this paper, we consider the Gauss quadrature formulae corresponding to some modifications of anyone of the four Chebyshev weights, considered by Gautschi and Li in \cite{gauli}. As it is well known, in the case of analytic integrands, the error of these quadrature formulas can be represented as a contour integral with a complex kernel.
Orive, Ramon +2 more
openaire +5 more sources
Measurement Error Estimation for Distributed Smart Meters Through a Modified BP Neural Network
Frontiers in Energy Research, 2022 Smart meters generally suffer degradation of metering accuracy and performance due to aging, faults, and other factors, which, however, are difficult to detect.
Tian Xia +5 more
doaj +1 more source
A derivation error that affects carbon balance models exists in the current implementation of the modified Arrhenius function [PDF]
New Phytologist, 2020 Summary Understanding biological temperature responses is crucial to predicting global carbon fluxes. The current approach to modelling temperature responses of photosynthetic capacity in large scale modelling efforts uses a modified Arrhenius equation.
Bridget Murphy, Joseph R. Stinziano
openaire +3 more sources
A Modified Error Function to Improve the Error Back-Propagation Algorithm for Multi-Layer Perceptrons [PDF]
ETRI Journal, 1995 This paper proposes a modified error function to improve the error back-propagation (EBP) algorithm for multi-Layer perceptrons (MLPs) which suffers from slow learning speed. It can also suppress over-specialization for training patterns that occurs in an algorithm based on a cross-entropy cost function which markedly reduces learning time.
Sang-Hoon Oh Oh, Youngjik Lee Lee
openaire +1 more source
Robust Liquid Level Control of Quadruple Tank System: A Nonlinear Model-Free Approach
Actuators, 2023 In this paper, two new versions of modified active disturbance rejection control (MADRC) are proposed to stabilize a nonlinear quadruple tank system and control the water levels of the lower two tanks in the presence of exogenous disturbances, parameter ...
Zahraa Sabah Hashim +13 more
doaj +1 more source
Ibn Al-Haitham Journal for Pure and Applied Sciences, 2019 In this paper, a Monte Carlo Simulation technique is used to compare the performance of MLE and the standard Bayes estimators of the reliability function of the one parameter exponential distribution.Two types of loss functions are adopted, namely,
Mohammed Jamel Ali +1 more
doaj +1 more source
Kernel Minimum Error Entropy Based Estimator for MIMO Radar in Non-Gaussian Clutter
IEEE Access, 2021 In this paper, a kernel minimum error entropy (KMEE) based estimator is proposed for the estimation of multiple targets’ direction of departure (DOD), the direction of arrival (DOA), and the Doppler shift with multiple input multiple output radar ...
Uday Kumar Singh +3 more
doaj +1 more source
Efficient estimation of Pareto model: Some modified percentile estimators. [PDF]
PLoS ONE, 2018 The article proposes three modified percentile estimators for parameter estimation of the Pareto distribution. These modifications are based on median, geometric mean and expectation of empirical cumulative distribution function of first-order statistic.
Sajjad Haider Bhatti +5 more
doaj +1 more source