We present a dynamic network rewiring (DNR) method to generate pruned deep neural (DNN) models that both are robust against adversarially generated images and maintain high accuracy on clean images. In particular, the disclosed DNR training is based unified constrained optimization formulation using novel hybrid loss function merges sparse learning with adversarial training. This strategy dynam...

For the edge e = uv of a graph G, let nu = n(u|G) be the number of vertices of G lying closer to the vertex u than to the vertex v and nv= n(v|G) can be defined simailarly. Then the ABCGG index of G is defined as ABC...

In this research, selective adsorption of U(VI) in aqueous solutionsin the presence of various lanthanide ions by using U(VI)-imprinted cryogel polymer was conducted. For this purpose, the prepared pHEMA-(MAH)3-U(VI) cryogel polymer by free radical polymerization method. U(VI) was desorbed with 5.0 mol/L HNO3 and thus U(VI)-imprinted were created onto p-HEMA-(...

the mechanistic aspects of the electrochemical reduction of u(v1) nitrate at a mercury electrode in chloroform, using three different supporting electrolytes, are described. in the presence of tetrabutylarnmonium perchlorate (tbap) as a neutral electrolyte, uo (no ) , reduced in two irreversible electrochemical steps to form a u(v1)-u(v) binuclear complex and u(iv), respectively. in the presenc...

In this paper‎, ‎we consider the following Kirchhoff-type equations‎: ‎$-‎left(a+bint_{mathbb{R}^{3}}|nabla u|^{2}right)Delta u+V(x) u=lambda$ $f(x,u)+u^{5}‎, ‎quad mbox{in }mathbb{R}^{3},$ ‎$u(x)>0‎, ‎quad mbox{in }mathbb{R}^{3},$ ‎$uin H^{1}(mathbb{R}^{3})‎ ,‎$ ‎ ‎‎‎where $a,b>0$ are constants and $lambda$ is a positive parameter‎. ‎The aim of this paper is to study the existence of positive ...

When visual sensor networks (VSNs) enter daily life in society, they not only bring great convenience but also cause people to worry about privacy. Traditional image encryption often displays the effect of snowflake encrypted images playing a part privacy protection, usability directly browsing taken by nodes VSNs is sacrificed. Recently, thumbnail-preserving (TPE) has been proposed balance and...

The reverse degree distance of a connected graph $G$ is defined in discrete mathematical chemistry as [ r (G)=2(n-1)md-sum_{uin V(G)}d_G(u)D_G(u), ] where $n$, $m$ and $d$ are the number of vertices, the number of edges and the diameter of $G$, respectively, $d_G(u)$ is the degree of vertex $u$, $D_G(u)$ is the sum of distance between vertex $u$ and all other vertices of $G$, and $V(G)$ is the...

let $r$ be a 2-torsion free ring and $u$ be a square closed lie ideal of $r$. suppose that $alpha, beta$ are automorphisms of $r$. an additive mapping $delta: r longrightarrow r$ is said to be a jordan left $(alpha,beta)$-derivation of $r$ if $delta(x^2)=alpha(x)delta(x)+beta(x)delta(x)$ holds for all $xin r$. in this paper it is established that if $r$ admits an additive mapping $g : rlongrigh...

The atom bond connectivity index of a graph is a new topological index was defined by E. Estrada as ABC(G)  uvE (dG(u) dG(v) 2) / dG(u)dG(v) , where G d ( u ) denotes degree of vertex u. In this paper we present some bounds of this new topological index.