Continuous classification of spatio-temporal data streams using liquid state machines

Authors:
Stefan Schliebs;Doug Hunt
Affiliations:
Auckland University of Technology, New Zealand;Auckland University of Technology, New Zealand
Venue:
ICONIP'12 Proceedings of the 19th international conference on Neural Information Processing - Volume Part IV
Year:
2012

Citing 6
Cited 0

Spiking Neuron Models: An Introduction

Spiking Neuron Models: An Introduction
Real-time computing without stable states: a new framework for neural computation based on perturbations

Neural Computation
Gesture spotting with body-worn inertial sensors to detect user activities

Pattern Recognition
2008 Special Issue: Compact hardware liquid state machines on FPGA for real-time speech recognition

Neural Networks
Integrated feature and parameter optimization for an evolving spiking neural network

ICONIP'08 Proceedings of the 15th international conference on Advances in neuro-information processing - Volume Part I
Constructing robust liquid state machines to process highly variable data streams

ICANN'12 Proceedings of the 22nd international conference on Artificial Neural Networks and Machine Learning - Volume Part I

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes to use a Liquid State Machine (LSM) to classify inertial sensor data collected from horse riders into activities of interest. LSM was shown to be an effective classifier for spatio-temporal data and efficient hardware implementations on custom chips have been presented in literature that would enable relative easy integration into wearable technologies. We explore here the general method of applying LSM technology to domain constrained activity recognition using a synthetic data set. The aim of this study is to provide a proof of concept illustrating the applicability of LSM for the chosen problem domain.