Paper Info
Title
Identifying qualitatively different outcomes of actions: gaining autonomy through learning
Abstract
This paper presents an unsupervised method for learningmodels of environmental dynamics based on clusteringmultivariate time series. Experiments with aPioneer-1 mobile robot demonstrate the utility of themethod and show that the models acquired by the robotcorrelate surprisingly well with human models of the environment.Individual time series are obtained by recording theoutput of a subset of an agent's sensors. We call thesetime series experiences. An example of a sensor subseton the ...
Year
DOI
Venue
2000
10.1145/336595.337068
Agents
Keywords
Field
DocType
qualitatively different outcome,intelligent agents,fuzzy logic,mobile robot,neural networks,time series,software agents,genetic algorithms,javabeans
Intelligent agent,Computer science,Autonomy,Fuzzy logic,Software agent,JavaBeans,Artificial intelligence,Artificial neural network,Machine learning,Genetic algorithm
Conference
ISBN
Citations 
PageRank 
1-58113-230-1
6
0.77
References 
Authors
6
3
Name
Order
Citations
PageRank
Tim Oates11069190.77
Matthew D. Schmill29814.67
paul r cohen31927460.49