Maximum Likelihood Methods in Unknown Latent Variables

In this problem, the goal is to compute maximum-likelihood estimates when the observations can be viewed as incomplete data.

Parameters

  • nn: number of observations in sample
  • rr: number of parameters + latent variables

Filters

Computational Model

Randomization

Approximation

Algorithms Table

Displaying 7 of 7 algorithms

See more
Shaban; Amirreza; Mehrdad; Farajtabar2015O(n2log2n)O(n^2 \log^2 n)O(kd+d3)O(kd+d^3)
α-EM Algorithm2003O(n3)O(n^3)O(n+r)O(n+r)
Parameter-expanded expectation maximization (PX-EM)1998O(n3)O(n^3)O(n+r)O(n+r)
EM with Quasi-Newton Methods (Jamshidian; Mortaza; Jennrich; Robert I.)1997O(n2log3n)O(n^2 \log^3 n)O(n+r2)O(n+r^2)
Expectation conditional maximization either (ECME) (Liu; Chuanhai; Rubin; Donald B)1994O(n3)O(n^3)O(n+r)O(n+r)
Expectation conditional maximization (ECM)1993O(n3)O(n^3)O(n+r)O(n+r)
Expectation-Maximization (EM) algorithm1977O(n3)O(n^3)O(n+r)O(n+r)

Reductions Table

Insuffient Data to display table

Other relevant algorithms

Insuffient Data to display table