Robust Linear Neural Network for Constrained Quadratic Optimization
نویسندگان
چکیده
منابع مشابه
Supplement: Distributed Box-constrained Quadratic Optimization for Dual Linear SVM
In this section, we will establish a detailed convergence analysis for our methods. The main idea follows from the framework of Tseng & Yun (2009) and substitutes the local error bound with the global error bound given in Wang & Lin (2014). Note that the result in Yun (2014) does not directly apply here because H = I is assumed in that work, but their analysis can also be modified for other sym...
متن کاملDistributed Box-Constrained Quadratic Optimization for Dual Linear SVM
Training machine learning models sometimes needs to be done on large amounts of data that exceed the capacity of a single machine, motivating recent works on developing algorithms that train in a distributed fashion. This paper proposes an efficient box-constrained quadratic optimization algorithm for distributedly training linear support vector machines (SVMs) with large data. Our key technica...
متن کاملA Robust Knapsack Based Constrained Portfolio Optimization
Many portfolio optimization problems deal with allocation of assets which carry a relatively high market price. Therefore, it is necessary to determine the integer value of assets when we deal with portfolio optimization. In addition, one of the main concerns with most portfolio optimization is associated with the type of constraints considered in different models. In many cases, the resulted p...
متن کاملA Nonfeasible Quadratic Approximation Recurrent Neural Network for Equality Constrained Optimization Problems
Convex optimization techniques are widely used in the design and analysis of communication systems and signal processing algorithms. In this paper a novel recurrent neural network is presented for solving nonlinear strongly convex equality constrained optimization problems. The proposed neural network is based on recursive quadratic programming for nonlinear optimization, in conjunction with ho...
متن کاملAnother Simple Recurrent Neural Network for Quadratic and Linear Programming
A new recurrent neural network is proposed for solving quadratic and linear programming problems, which is derived from two salient existing neural networks. One of the predecessors has lower structural complexity but were not shown to be capable of solving degenerate QP problems including LP problems while the other does not have this limitation but has higer structural complexity. The propose...
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: Discrete Dynamics in Nature and Society
سال: 2017
ISSN: 1026-0226,1607-887X
DOI: 10.1155/2017/5073640