We propose two algorithms for solving global optimization problems on a hyperrectangle with an objective function satisfying the Vanderbei condition (this is also called $$\varepsilon $$ -Lipschitz continuous function). The belong to family of non-uniform covering methods. For we prove propositions about convergence -solution in terms function. illustrate performance using several test numerica...

Consider the Cauchy problem for an ordinary diierential equation _ x = g(t; x); x(0) = x; t 2 0; T]: (1:1) When g is continuous, the local existence of solutions is provided by Peano's theorem. Several existence and uniqueness results are known also in the case of a discontinuous right hand side 7]. We recall here the classical theorem of Carath eodory 8]: Theorem A. Let g : 0; T] IR n 7 ! IR n...

The ε-pseudospectral abscissa αε and radius ρε of an n× n matrix are, respectively, the maximal real part and the maximal modulus of points in its ε-pseudospectrum, defined using the spectral norm. It was proved in [A.S. Lewis and C.H.J. Pang, SIAM J. Optim., 19 (2008), pp. 1048–1072] that for fixed ε > 0, αε and ρε are Lipschitz continuous at a matrix A except when αε and ρε are attained at a ...

A new monotonicity, M-monotonicity, is introduced, and the resolvant operator of an M-monotone operator is proved to be single-valued and Lipschitz continuous. With the help of the resolvant operator, the positively semidefinite general variational inequality (VI) problem VI (S+,F +G) is transformed into a fixed point problem of a nonexpansive mapping. And a proximal point algorithm is construc...

We study the L2-time regularity of the Z-component of a Markovian BSDE, whose terminal condition is a function g of a forward SDE (Xt)0≤t≤T . When g is Lipschitz continuous, Zhang [1] proved that the related squared L2-time regularity is of order one with respect to the size of the time mesh. We extend this type of result to any function g, including irregular functions such as indicator functi...

This paper is concerned with the study of a penalization-gradient algorithm for solving variational inequalities, namely, find x ∈ C such that 〈Ax, y − x〉 ≥ 0 for all y ∈ C, where A : H → H is a single-valued operator, C is a closed convex set of a real Hilbert space H. Given Ψ : H → ∪ { ∞} which acts as a penalization function with respect to the constraint x ∈ C, and a penalization parameter ...

We derive two theorems combining existence with necessary conditions for the relaxed unilateral problem of the optimal control of ordinary differential equations in which the functions that define the problem are Lipschitz-continuous in the state variables. These theorems generalize the results presented in a previous paper [8] by the addition of unilateral constraints on the state and control ...

This paper constructs translation-invariant operators on L2.Rd /, which are Lipschitz-continuous to the action of diffeomorphisms. A scattering propagator is a path-ordered product of nonlinear and noncommuting operators, each of which computes the modulus of a wavelet transform. A local integration defines a windowed scattering transform, which is proved to be Lipschitz-continuous to the actio...

We study online optimization of smoothed piecewise constant functions over the domain [0, 1). This is motivated by the problem of adaptively picking parameters of learning algorithms as in the recently introduced framework by Gupta and Roughgarden (2016). Majority of the machine learning literature has focused on Lipschitz-continuous functions, or functions with bounded gradients. This is with ...

(1.1) ∂tb(x, u) + div(F̄ (t, x, u)− k∇u) = f(t, x, u, s), s(t, x) = ∫ t 0 K(t, z)ψ(u(z, x))dz in Ω × (0, T ], T < ∞, Ω ⊂ R is a bounded domain, ∂Ω ∈ C, see [26]. If Ω is convex, then ∂Ω is assumed to be Lipschitz continuous. We consider a Dirichlet boundary condition (1.2) u(t, x) = 0 on I × ∂Ω, I = (0, T ], together with the initial condition (1.3) u(0, x) = u0(x) x ∈ Ω. We assume 0 < ε ≤ ∂sb(x...