Data-driven MCMC for learning and inference in switching linear dynamic systems

Sang Min Oh, James M. Rehg, Tucker Balch, Frank Dellaert

Research output: Contribution to conferencePaperpeer-review

Abstract

Switching Linear Dynamic System (SLDS) models are a popular technique for modeling complex nonlinear dynamic systems. An SLDS has significantly more descriptive power than an HMM, but inference in SLDS models is computationally intractable. This paper describes a novel inference algorithm for SLDS models based on the Data-Driven MCMC paradigm. We describe a new proposal distribution which substantially increases the convergence speed. Comparisons to standard deterministic approximation methods demonstrate the improved accuracy of our new approach. We apply our approach to the problem of learning an SLDS model of the bee dance. Honeybees communicate the location and distance to food sources through a dance that takes place within the hive. We learn SLDS model parameters from tracking data which is automatically extracted from video. We then demonstrate the ability to successfully segment novel bee dances into their constituent parts, effectively decoding the dance of the bees.

Original languageEnglish (US)
Pages944-949
Number of pages6
StatePublished - 2005
Externally publishedYes
Event20th National Conference on Artificial Intelligence and the 17th Innovative Applications of Artificial Intelligence Conference, AAAI-05/IAAI-05 - Pittsburgh, PA, United States
Duration: Jul 9 2005Jul 13 2005

Other

Other20th National Conference on Artificial Intelligence and the 17th Innovative Applications of Artificial Intelligence Conference, AAAI-05/IAAI-05
Country/TerritoryUnited States
CityPittsburgh, PA
Period7/9/057/13/05

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Data-driven MCMC for learning and inference in switching linear dynamic systems'. Together they form a unique fingerprint.

Cite this