论文阅读 [TPAMI-2022] A Fast Binary Quadratic Programming Solver Based on Stochastic Neighborhood Search

论文阅读 [TPAMI-2022] A Fast Binary Quadratic Programming Solver Based on Stochastic Neighborhood Search

论文搜索(studyai.com)

搜索论文: A Fast Binary Quadratic Programming Solver Based on Stochastic Neighborhood Search

搜索论文: http://www.studyai.com/search/whole-site/?q=A+Fast+Binary+Quadratic+Programming+Solver+Based+on+Stochastic+Neighborhood+Search

关键字(Keywords)

Search problems; Quadratic programming; Pattern recognition; Computational complexity; Image restoration; Binary quadratic programming; stochastic optimization; binary restoration; graph bisection; optimization

机器视觉; 运筹与优化

随机优化; 图像/视频修复

摘要(Abstract)

Many image processing and pattern recognition problems can be formulated as binary quadratic programming (BQP) problems.

许多图像处理和模式识别问题可以被表述为二元二次规划（BQP）问题。

However, solving a large BQP problem with a good quality solution and low computational time is still a challenging unsolved problem.

然而，以高质量的解决方案和低计算时间解决大型BQP问题仍然是一个具有挑战性的未解决的问题。

Current methodologies either adopt an independent random search in a semi-definite space or perform search in a relaxed biconvex space.

目前的方法学要么在半无限空间中采用独立的随机搜索，要么在一个宽松的双凸空间中进行搜索。

However, the independent search has great computation cost as many different trials are needed to get a good solution.

然而，独立搜索有很大的计算成本，因为需要许多不同的试验来获得一个好的解决方案。

The biconvex search only searches the solution in a local convex ball, which can be a local optimal solution.

双凸搜索只在局部凸球中搜索解决方案，这可以是一个局部最优解决方案。

In this paper, we propose a BQP solver that alternatingly applies a deterministic search and a stochastic neighborhood search.

在本文中，我们提出了一个BQP求解器，它交替应用确定性搜索和随机邻域搜索。

The deterministic search iteratively improves the solution quality until it satisfies the KKT optimality conditions.

确定性搜索迭代地提高解决方案的质量，直到它满足KKT优化条件。

The stochastic search performs bootstrapping sampling to the objective function constructed from the potential solution to find a stochastic neighborhood vector.

随机搜索对由潜在解决方案构建的目标函数进行引导抽样，以找到随机邻域向量。

These two steps are repeated until the obtained solution is better than many of its stochastic neighborhood vectors.

这两个步骤重复进行，直到获得的解决方案优于其许多随机邻域向量。

We compare the proposed solver with several state-of-the-art methods for a range of image processing and pattern recognition problems.

我们将所提出的求解器与一些最先进的方法进行比较，以解决一系列的图像处理和模式识别问题。

Experimental results showed that the proposed solver not only outperformed them in solution quality but also with the lowest computational complexity…

实验结果表明，所提出的解算器不仅在解算质量上优于他们，而且计算复杂度也最低。

作者(Authors)

[‘Benson Shu Lam’, ‘Alan Wee-Chung Liew’]