Learning Hierarchical Models of Activity

Abstract

This paper investigates learning hierarchical statistical activity models in indoor environments. The Abstract Hidden Markov Model (AHMM) is used to represent behaviors in stochastic environments. We train the model using both labeled and unlabeled data and estimate the parameters using Expectation Maximization (EM). Results are shown on three datasets: data collected in lab, entryway, and home environments. The results show that hierarchical models outperform flat models.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2005
Accession Number
ADA440281

Entities

People

  • Sarah Osentoski
  • Sridhar Mahadevan
  • Victoria Manfred

Organizations

  • University of Massachusetts Amherst

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Bayesian Networks
  • Classification
  • Computer Science
  • Covariance
  • Data Sets
  • Environment
  • Hidden Markov Models
  • Hierarchies
  • Instructions
  • Learning
  • Markov Models
  • Models
  • Observation
  • Probability
  • Probability Distributions

Fields of Study

  • Computer science

Readers

  • Computational Modeling and Simulation
  • Neural Network Machine Learning.
  • Nuclear Civil Defense.