全球时空更新

Borealis -布尔优化问题的一个广义全局更新算法

borealis - A generalized global update algorithm for Boolean optimization problems

非确定性多项式（NP）类布尔变量优化问题在计算机科学、数学、物理和工业应用中具有重要意义。最值得注意的是，解决约束满足问题，这是相关的自旋玻璃类哈密顿在物理学中，仍然是一个困难的数值任务。因此，人们对设计有效的算法来解决这些计算困难的问题产生了极大的兴趣。受并行回火蒙特卡罗方法的启发，结合伊辛自旋玻璃的无排斥等能簇算法，我们提出了一个广义的全局更新优化启发式算法，可应用于不同的NP完全问题与布尔变量。全局聚类更新允许相空间的广泛采样，从而大大加快优化。通过仔细调整伪温度（需要随机化的配置）的问题，我们表明，该方法可以有效地解决优化问题的过度约束或拓扑结构具有较大的网站渗透阈值。我们说明了效率的启发式范式优化问题，如最大可满足性问题和顶点覆盖问题。

GPU-based Ising Computing for Solving Balanced Min-Cut Graph Partitioning Problem

Ising computing provides a new computing paradigm for many hard combinatorial optimization problems. Ising computing essentially tries to solve the quadratic unconstrained binary optimization problem, which is also described by the Ising spin glass model and is also the basis for so-called Quantum Annealing computers. In this work, we propose a novel General Purpose Graphics Processing Unit (GPGPU) solver for the balanced min-cut graph partitioning problem, which has many applications in the area of design automation and others. Ising model solvers for the balanced min-cut partitioning problem…

Time evolution of an infinite projected entangled pair state: a gradient tensor update in the tangent space

Time evolution of an infinite 2D many body quantum lattice system can be described by the Suzuki-Trotter decomposition applied to the infinite projected entangled pair state (iPEPS). Each Trotter gate increases the bond dimension of the tensor network, $D$, that has to be truncated back in a way that minimizes a suitable error measure. This paper goes beyond simplified error measures -- like the one used in the full update (FU), the simple update (SU), and their intermediate neighborhood tensor update (NTU) -- and directly maximizes an overlap between the exact iPEPS with the increased bond di…