A Caltech Library Service

The Identification of Discrete Mixture Models


Gordon, Spencer Lane (2023) The Identification of Discrete Mixture Models. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/ebf5-0b48.


In this thesis we discuss a variety of results on learning and identifying discrete mixture models, i.e., distributions that are a convex combination of k from a known class C of distributions. We first consider the case where C is the class of binomial distributions, before generalizing to the case of product distributions. We provide a necessary condition for identifiability of mixture of products distributions as well as a generalization to structured mixtures over multiple latent variables.

Item Type:Thesis (Dissertation (Ph.D.))
Subject Keywords:Mixture Models; Identifiability; Algorithms; Complexity; Statistical Learning Theory
Degree Grantor:California Institute of Technology
Division:Engineering and Applied Science
Major Option:Computer Science
Thesis Availability:Public (worldwide access)
Research Advisor(s):
  • Vidick, Thomas G. (co-advisor)
  • Schulman, Leonard J. (co-advisor)
Thesis Committee:
  • Yue, Yisong (chair)
  • Vidick, Thomas Georges
  • Schulman, Leonard J.
  • Wierman, Adam C.
Defense Date:19 January 2023
Record Number:CaltechTHESIS:02072023-112938936
Persistent URL:
Related URLs:
URLURL TypeDescription adapted for Chapter 1
Gordon, Spencer Lane0000-0002-7101-2370
Default Usage Policy:No commercial reproduction, distribution, display or performance rights in this work are provided.
ID Code:15101
Deposited By: Spencer Gordon
Deposited On:17 Feb 2023 17:37
Last Modified:26 Apr 2023 21:21

Thesis Files

[img] PDF - Final Version
See Usage Policy.


Repository Staff Only: item control page