Rotations in Curved Trajectories for Unconstrained Minimization
نویسنده
چکیده
Curved Trajectories Algorithm (CTA) is a package for the minimization of unconstrained functions of several variables with intervals on the variables. The core algorithm is novel in that steps may follow polynomial space curves instead of straight lines. The space curves result from truncations of a Taylor series expansion of the Gradient inverse function. When the series is convergent and the current guess of the solution is far, appropriate rotations of the space curve may allow further progress at a step. Improvements are significant for some functions, reducing the number of Hessians required or improving the accuracy of the solution. This idea may be useful to other minimization packages.
منابع مشابه
A variable order algorithm for unconstrained minimization
A variable order algorithm is proposed for the rini^ization of a function cff several variables. This algorithm has an order of convergence *.s high as four, good accuracy, and features a scalar subproblem at each iteration which may be along curved trajectories in the space of the independent variables. Numerical results en five standard test problems indicate superior performance when compare...
متن کاملDifferences in Context and Feedback Result in Different Trajectories and Adaptation Strategies in Reaching
Computational models of motor control have often explained the straightness of horizontal planar reaching movements as a consequence of optimal control. Departure from rectilinearity is thus regarded as sub-optimal. Here we examine if subjects may instead select to make curved trajectories following adaptation to force fields and visuomotor rotations. Separate subjects adapted to force fields w...
متن کاملA NEW APPROACH TO THE SOLUTION OF SENSITIVITY MINIMIZATION IN LINEAR STATE FEEDBACK CONTROL
In this paper, it is shown that by exploiting the explicit parametric state feedback solution, it is feasible to obtain the ultimate solution to minimum sensitivity problem. A numerical algorithm for construction of a robust state feedback in eigenvalue assignment problem for a controllable linear system is presented. By using a generalized parametric vector companion form, the problem of eigen...
متن کاملBlock Transform Adaptation by Stochastic Gradient Descent
The problem of computing the eigendecomposition of an N N symmetric matrix is cast as an unconstrained minimization of either of two performance measures. The K = N(N 1)=2 independent parameters represent angles of distinct Givens rotations. Gradient descent is applied to the minimization problem, step size bounds for local convergence are given, and similarities to LMS adaptive filtering are n...
متن کاملA class of derivative-free nonmonotone optimization algorithms employing coordinate rotations and gradient approximations
In this paper we study a class of derivative-free unconstrained minimization algorithms employing nonmonotone inexact linesearch techniques along a set of suitable search directions. In particular, we define globally convergent nonmonotone versions of some well-known derivativefree methods and we propose a new algorithm combining coordinate rotations with approximate simplex gradients. Through ...
متن کامل