Better Approximate Inference for Partial Likelihood Models with a Latent Structure

10/22/2019
by   Amrith Setlur, et al.
0

Temporal Point Processes (TPP) with partial likelihoods involving a latent structure often entail an intractable marginalization, thus making inference hard. We propose a novel approach to Maximum Likelihood Estimation (MLE) involving approximate inference over the latent variables by minimizing a tight upper bound on the approximation gap. Given a discrete latent variable Z, the proposed approximation reduces inference complexity from O(|Z|^c) to O(|Z|). We use convex conjugates to determine this upper bound in a closed form and show that its addition to the optimization objective results in improved results for models assuming proportional hazards as in Survival Analysis.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
12/06/2019

A new inequality for maximum likelihood estimation in statistical models with latent variables

Maximum-likelihood estimation (MLE) is arguably the most important tool ...
research
07/13/2021

Graphical Laplace-approximated maximum likelihood estimation: approximated likelihood inference for network data analysis

We derive Laplace-approximated maximum likelihood estimators (GLAMLEs) o...
research
06/02/2023

A Stochastic Multivariate Latent Variable Model For Categorical Responses

This paper introduces a mathematical framework of a stochastic process m...
research
05/23/2018

Learning latent variable structured prediction models with Gaussian perturbations

The standard margin-based structured prediction commonly uses a maximum ...
research
11/25/2019

Analysis of odds, probability, and hazard ratios: From 2 by 2 tables to two-sample survival data

Analysis of 2 by 2 tables and two-sample survival data has been widely u...
research
09/28/2018

Variational Bayesian Inference for Audio-Visual Tracking of Multiple Speakers

In this paper we address the problem of tracking multiple speakers via t...
research
03/04/2015

Bethe Learning of Conditional Random Fields via MAP Decoding

Many machine learning tasks can be formulated in terms of predicting str...

Please sign up or login with your details

Forgot password? Click here to reset