Learning to Reproduce Fluctuating Time Series by Inferring Their Time-Dependent Stochastic Properties: Application in Robot Learning Via Tutoring

Cited 21 time in webofscience Cited 0 time in scopus
  • Hit : 276
  • Download : 3
This study proposes a novel type of dynamic neural network model that can learn to extract stochastic or fluctuating structures hidden in time series data. The network learns to predict not only the mean of the next input state, but also its time-dependent variance. The training method is based on maximum likelihood estimation by using the gradient descent method and the likelihood function is expressed as a function of the estimated variance. Regarding the model evaluation, we present numerical experiments in which training data were generated in different ways utilizing Gaussian noise. Our analysis showed that the network can predict the time-dependent variance and the mean and it can also reproduce the target stochastic sequence data by utilizing the estimated variance. Furthermore, it was shown that a humanoid robot using the proposed network can learn to reproduce latent stochastic structures hidden in fluctuating tutoring trajectories. This learning scheme is essential for the acquisition of sensory-guided skilled behavior.
Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
Issue Date
2013-12
Language
English
Article Type
Article
Keywords

RECURRENT NEURAL NETWORKS; DYNAMICAL-SYSTEMS; HUMANOID ROBOT; MODEL; TASK

Citation

IEEE TRANSACTIONS ON AUTONOMOUS MENTAL DEVELOPMENT, v.5, no.4, pp.298 - 310

ISSN
1943-0604
DOI
10.1109/TAMD.2013.2258019
URI
http://hdl.handle.net/10203/187432
Appears in Collection
EE-Journal Papers(저널논문)
Files in This Item
This item is cited by other documents in WoS
⊙ Detail Information in WoSⓡ Click to see webofscience_button
⊙ Cited 21 items in WoS Click to see citing articles in records_button

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0