Activity recognition with time-delay embeddings

Jordan Frank, Shie Mannor, Doina Precup

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We outline an approach that uses time-delay embedding models and machine learning in order to recognize physical activities, based on data coming from cell phone accelerometers. The approach is very robust, cheap in terms of the amount of data and computation required, and easy to deploy.

Original languageEnglish
Title of host publicationComputational Physiology - Papers from the AAAI Spring Symposium, Technical Report
Pages13-14
Number of pages2
StatePublished - 2011
Event2011 AAAI Spring Symposium - Stanford, CA, United States
Duration: 21 Mar 201123 Mar 2011

Publication series

NameAAAI Spring Symposium - Technical Report
VolumeSS-11-04

Conference

Conference2011 AAAI Spring Symposium
Country/TerritoryUnited States
CityStanford, CA
Period21/03/1123/03/11

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Activity recognition with time-delay embeddings'. Together they form a unique fingerprint.

Cite this