logo
    Improved Hessian estimation for adaptive random directions stochastic approximation
    4
    Citation
    10
    Reference
    10
    Related Paper
    Citation Trend
    Abstract:
    We propose an improved Hessian estimation scheme for the second-order random directions stochastic approximation (2RDSA) algorithm [1]. The proposed scheme, inspired by [2], reduces the error in the Hessian estimate by (i) incorporating a zero-mean feedback term; and (ii) optimizing the step-sizes used in the Hessian recursion of 2RDSA.We prove that 2RDSA with our Hessian improvement scheme (2RDSA-IH) converges asymptotically to the true Hessian. The advantage with 2RDSA-IH is that it requires only 75% of the simulation cost per-iteration for 2SPSA with improved Hessian estimation (2SPSA-IH) [2]. Numerical experiments show that 2RDSA-IH outperforms both 2SPSA-IH and 2RDSA without the improved Hessian estimation scheme.
    Keywords:
    Hessian matrix
    Hessian equation
    Stochastic Approximation
    In this short note we present new integral formulas for the Hessian determinant. We use them for new definitions of Hessian under minimal regularity assumptions. The Hessian becomes a continuous linear functional on a Sobolev space.
    Hessian matrix
    Hessian equation
    Sobolev inequality
    Citations (0)
    We show that the second–order condition for strict local extrema in both constrained and unconstrained optimization problems can be expressed solely in terms of principal minors of the (Lagrengean) Hessian. This approach unifies the determinantal tests in the sense that the second-order condition can be always given solely in terms of Hessian matrix.
    Hessian matrix
    Hessian equation
    Maxima and minima
    Matrix (chemical analysis)
    Citations (7)
    We derive Hessian estimates for convex solutions to quadratic Hessian equation by a compactness argument.
    Hessian matrix
    Hessian equation
    Argument (complex analysis)
    Citations (0)
    The $k$-Hessian operator $\sigma_k$ is the $k$-th elementary symmetric function of the eigenvalues of the Hessian. It is known that the $k$-Hessian equation $\sigma_k(D^2u)=f$ with Dirichlet boundary condition $u=0$ is variational; indeed, this problem can be studied by means of the $k$-Hessian energy $-\int u\sigma_k(D^2u)$. We construct a natural boundary functional which, when added to the $k$-Hessian energy, yields as its critical points solutions of $k$-Hessian equations with general non-vanishing boundary data. As a consequence, we prove a sharp Sobolev trace inequality for $k$-admissible functions $u$ which estimates the $k$-Hessian energy in terms of the boundary values of $u$.
    Hessian equation
    Hessian matrix
    TRACE (psycholinguistics)
    Sobolev inequality
    Sigma
    Citations (0)
    In this paper, we consider the following system coupled by multiparameter k ‐Hessian equations Here, λ 1 and λ 2 are positive parameters, Ω is the unit ball in R n , S k ( D 2 u ) is the k ‐Hessian operator of u , . Applying the eigenvalue theory in cones, several new results are obtained for the existence and multiplicity of nontrivial radial solutions for the above k ‐Hessian system. In particular, we study the dependence of the nontrivial radial solution on the parameter λ 1 and λ 2 . This is probably the first time that a system of equations, especially with fully nonlinear equations, has been studied by applying this technique. Finally, as an application, we obtain sufficient conditions for the existence of nontrivial radial solutions of the power‐type coupled system of k ‐Hessian equations, which is new even for the special case k =1 and extends a previous result for the case k = n .
    Hessian equation
    Hessian matrix
    Multiplicity (mathematics)
    Unit sphere
    Ball (mathematics)
    Citations (9)
    The $k$-Hessian operator $\sigma_k$ is the $k$-th elementary symmetric function of the eigenvalues of the Hessian. It is known that the $k$-Hessian equation $\sigma_k(D^2u)=f$ with Dirichlet boundary condition $u=0$ is variational; indeed, this problem can be studied by means of the $k$-Hessian energy $-\int u\sigma_k(D^2u)$. We construct a natural boundary functional which, when added to the $k$-Hessian energy, yields as its critical points solutions of $k$-Hessian equations with general non-vanishing boundary data. As a consequence, we prove a sharp Sobolev trace inequality for $k$-admissible functions $u$ which estimates the $k$-Hessian energy in terms of the boundary values of $u$.
    Hessian equation
    Hessian matrix
    TRACE (psycholinguistics)
    Sobolev inequality
    Sigma
    Citations (1)