Maximum likelihood estimation (MLE) is one of the most important methods inmachine learning, and the expectation-maximization (EM) algorithm is often usedto obtain maximum likelihood estimates. However, EM heavily depends on initialconfigurations and fails to find the global optimum. On the other hand, in thefield of physics, quantum annealing (QA) was proposed as a novel optimizationapproach. Motivated by QA, we propose a quantum annealing extension of EM,which we call the deterministic quantum annealing expectation-maximization(DQAEM) algorithm. We also discuss its advantage in terms of the path integralformulation. Furthermore, by employing numerical simulations, we illustrate howit works in MLE and show that DQAEM outperforms EM.
展开▼
