Estimating Shape Parameters of Piecewise Linear-Quadratic Problems
نویسندگان
چکیده
Piecewise Linear-Quadratic (PLQ) penalties are widely used to develop models in statistical inference, signal processing, and machine learning. Common examples of PLQ include least squares, Huber, Vapnik, 1-norm, their asymmetric generalizations. Properties these estimators depend on the choice penalty its shape parameters, such as degree asymmetry for quantile loss, transition point between linear quadratic pieces Huber function.
منابع مشابه
Piecewise-Linear Approximation of Quadratic Problems
This report covers the application of a piecewise-linear approximation to a quadratic program, QP. In this case the program is one that solves 3D conformal radiation treatment problem and returns optimal beam weights. Four different methods of approximating the quadratic problem were tested and compared to each other and the original problem. The methods are δ method, λ method, σ method, and γ ...
متن کاملAchieving Linear or Quadratic Convergence on Piecewise Smooth Optimization Problems
Many problems in machine learning involve objective functions that are piecewise smooth [7] due to the occurrence of absolute values mins and maxes in their evaluation procedures. See e.g. [8]. For such function we derived in [3] first order (KKT) and second order (SSC) optimality conditions, which can be checked on the basis of a local piecewise linearization [2] that can be computed in an AD ...
متن کاملPiecewise linear quadratic optimal control
The use of piecewise quadratic cost functions is extended from stability analysis of piecewise linear systems to performance analysis and optimal control. Lower bounds on the optimal control cost are obtained by semi-definite programming based on the Bellman inequality. This also gives an approximation to the optimal control law. An upper bound to the optimal cost is obtained by another convex ...
متن کاملPiecewise Linear Solution Paths for Parametric Piecewise Quadratic Programs
Recently, pathfollowing algorithms for parametric optimization problems with piecewise linear solution paths have been developed within the field of regularized regression. This paper presents a generalization of these algorithms to a wider class of problems, namely a class of parametric piecewise quadratic programs and related problems. By using pathfollowing algorithms that exploit the piecew...
متن کاملNonconvex piecewise linear knapsack problems
This paper considers the minimization version of a class of nonconvex knapsack problems with piecewise linear cost structure. The items to be included in the knapsack have a divisible quantity and a cost function. An item can be included partially in the given quantity range and the cost is a nonconvex piecewise linear function of quantity. Given a demand, the optimization problem is to choose ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Open journal of mathematical optimization
سال: 2021
ISSN: ['2777-5860']
DOI: https://doi.org/10.5802/ojmo.10