Identifying Linear Combinations of Ridge Functions
نویسندگان
چکیده
This paper is about an inverse problem. We assume we are given a function f(x) which is some sum of ridge functions of the form ∑m i=1 gi(a i · x) and we just know an upper bound on m. We seek to identify the functions gi and also the directions a i from such limited information. Several ways to solve this nonlinear problem are discussed in this work. §
منابع مشابه
ارزیابی پویشگر ریسک به منظور شناسایی ریسکهای در حال ظهور با استفاده از مدل آنالیز تشدید کارکرد: مطالعهی موردی در یک واحد فرایندی
Background and aim: Today, it was revealed that Socio-technical systems did not have a bimodal nature and interactions in these systems are complex and non-linear. Consequently, since risks can be emerged as non-linear combinations of performance variability, so traditional methods of risk assessment are not able to capture these combinations. The present paper is aimed at identifying the eme...
متن کاملSmoothness and uniqueness in ridge function representation
In this note we consider problems of uniqueness, smoothness and representation of linear combinations of a finite number of ridge functions with fixed directions. c ⃝ 2012 Royal Dutch Mathematical Society (KWG). Published by Elsevier B.V. All rights reserved.
متن کاملInterpolation on lines by ridge functions
In this paper we study the problem of interpolation on straight lines by linear combinations of a finite number of ridge functions with fixed directions. We analyze in totality the case of one and two directions. c ⃝ 2013 Elsevier Inc. All rights reserved.
متن کاملUniform Approximation by Neural Networks Activated by First and Second Order Ridge Splines
We establish sup-norm error bounds for functions that are approximated by linear combinations of first and second order ridge splines and show that these bounds are near-optimal.
متن کاملUniversal Approximation by Ridge Computational Models and Neural Networks: A Survey
Computational models made up of linear combinations of ridge basis functions, widely used in machine learning and artificial intelligence, are considered. For such models, the literature on the so-called “universal approximation property” is surveyed. Different approaches, proof techniques, and tools are examined.
متن کامل