Results 1 to 10 of about 5,939,728 (334)
Smooth Function Approximation by Deep Neural Networks with General Activation Functions [PDF]
Entropy, 2019 There has been a growing interest in expressivity of deep neural networks. However, most of the existing work about this topic focuses only on the specific activation function such as ReLU or sigmoid.
Ilsang Ohn, Yongdai Kim
doaj +2 more sources
Function approximation method based on weights gradient descent in reinforcement learning
网络与信息安全学报, 2023 Function approximation has gained significant attention in reinforcement learning research as it effectively addresses problems with large-scale, continuous state, and action space.Although the function approximation algorithm based on gradient descent ...
Xiaoyan QIN +3 more
doaj +3 more sources
Arbitrarily Accurate Analytical Approximations for the Error Function
Mathematical and Computational Applications, 2022 A spline-based integral approximation is utilized to define a sequence of approximations to the error function that converge at a significantly faster manner than the default Taylor series.
Roy M. Howard
doaj +1 more source
Approximation of hysteresis functional
Journal of Computational and Applied Mathematics, 2021 We develop a practical discrete model of hysteresis based on nonlinear play and generalized play, for use in first-order conservation laws with applications to adsorption-desorption hysteresis models. The model is easy to calibrate from sparse data, and offers rich secondary curves. We compare it with discrete regularized Preisach models. We also prove
Malgorzata Peszynska, Ralph E. Showalter
openaire +4 more sources
Schröder-Based Inverse Function Approximation
Axioms, 2023 Schröder approximations of the first kind, modified for the inverse function approximation case, are utilized to establish general analytical approximation forms for an inverse function.
Roy M. Howard
doaj +1 more source
On approximately monotone and approximately Hölder functions [PDF]
Periodica Mathematica Hungarica, 2020 AbstractA real valued functionfdefined on a real open intervalIis called$$\Phi $$Φ-monotone if, for all$$x,y\in I$$x,y∈Iwith$$x\le y$$x≤yit satisfies$$\begin{aligned} f(x)\le f(y)+\Phi (y-x), \end{aligned}$$f(x)≤f(y)+Φ(y-x),where$$ \Phi :[0,\ell (I) [ \rightarrow \mathbb {R}_+$$Φ:[0,ℓ(I)[→R+is a given nonnegative error function, where$$\ell (I)$$ℓ(I ...
Angshuman R. Goswami, Zsolt Páles
openaire +3 more sources
Digital Fixed-Point Low Powered Area Efficient Function Estimation for Implantable Devices
IEEE Access, 2022 This article introduces a new multiplier-less 32-bit fixed point architecture for estimating complex non-linear functions based on adapted shift only series expansions.
James B. Romaine +2 more
doaj +1 more source
Using an Opportunity Matrix to Select Centers for RBF Neural Networks
Algorithms, 2023 When designed correctly, radial basis function (RBF) neural networks can approximate mathematical functions to any arbitrary degree of precision. Multilayer perceptron (MLP) neural networks are also universal function approximators, but RBF neural ...
Daniel S. Soper
doaj +1 more source
On the Universally Optimal Activation Function for a Class of Residual Neural Networks
AppliedMath, 2022 While non-linear activation functions play vital roles in artificial neural networks, it is generally unclear how the non-linearity can improve the quality of function approximations.
Feng Zhao, Shao-Lun Huang
doaj +1 more source
Mathematics, 2022 The paper considers the slope flow simulation and the problem of finding the optimal parameter values of this mathematical model. The slope flow is modeled using the finite volume method applied to the Reynolds-averaged Navier–Stokes equations with ...
Konstantin Barkalov +5 more
doaj +1 more source