论文标题
哈密顿交替的非凸优化
Non-Convex Optimization by Hamiltonian Alternation
论文作者
论文摘要
非凸优化的主要障碍是陷入本地最小值的问题。我们介绍了一种新颖的元神经来处理这个问题，创建了一种替代的哈密顿量，该哈密顿人仅在选定的能量范围内与原始的汉密尔顿人共享Minima。我们发现，按顺序将每个哈密顿量的每一个哈密顿量最小化允许算法逃脱局部最小值。当知道基态能量时，这种技术尤其简单，即使没有这些知识，也可以获得改进。我们通过使用它来找到Sherrington-Kirkpatrick旋转玻璃的实例来证明这一技术。
A major obstacle to non-convex optimization is the problem of getting stuck in local minima. We introduce a novel metaheuristic to handle this issue, creating an alternate Hamiltonian that shares minima with the original Hamiltonian only within a chosen energy range. We find that repeatedly minimizing each Hamiltonian in sequence allows an algorithm to escape local minima. This technique is particularly straightforward when the ground state energy is known, and one obtains an improvement even without this knowledge. We demonstrate this technique by using it to find the ground state for instances of a Sherrington-Kirkpatrick spin glass.
