Long short-term memory in recognizing behavior sequences on humanoid robot.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In order for robots to learn more complex behaviors, recognizing primitive behaviors plays a fundamental role. Research has shown that the recognition of primitive behaviors such as basic gestures enables robots to learn more complex behaviors as combinations of these simple, primitive behaviors. The focus of this study is to investigate the tolerance of neural network models to noisy inputs. We compare and evaluate several neural network architectures including the multilayer perceptron (MLP), time-delay neural network (TDNN), recurrent neural network (RNN) and the Long Short-Term Memory (LSTM). We show that the LSTM is superior to other models in terms of its robustness noisy inputs subjected to Gaussian noise.

Original languageEnglish
Title of host publicationProceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages859-866
Number of pages8
ISBN (Electronic)9781538626337
DOIs
Publication statusPublished - 15 May 2019
EventJoint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018 - Toyama, Japan
Duration: 05 Dec 201808 Dec 2018

Publication series

NameProceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018

Conference

ConferenceJoint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018
CountryJapan
CityToyama
Period05/12/1808/12/18

Fingerprint

Humanoid Robot
Memory Term
Robots
Neural networks
Recurrent neural networks
Multilayer neural networks
Network architecture
Robot
Neural Networks
Time delay
Gaussian Noise
Recurrent Neural Networks
Network Architecture
Gesture
Perceptron
Neural Network Model
Tolerance
Multilayer
Time Delay
Long short-term memory

All Science Journal Classification (ASJC) codes

  • Human-Computer Interaction
  • Logic
  • Artificial Intelligence
  • Computational Theory and Mathematics
  • Computer Science Applications
  • Theoretical Computer Science

Cite this

Dickson Neoh, T. H., Mohamed Sahari, K. S., & Loo, C. K. (2019). Long short-term memory in recognizing behavior sequences on humanoid robot. In Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018 (pp. 859-866). [8716108] (Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SCIS-ISIS.2018.00142
Dickson Neoh, Tze How ; Mohamed Sahari, Khairul Salleh ; Loo, Chu Kiong. / Long short-term memory in recognizing behavior sequences on humanoid robot. Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 859-866 (Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018).
@inproceedings{07c7c4e271394798a480dc58fe31ebea,
title = "Long short-term memory in recognizing behavior sequences on humanoid robot.",
abstract = "In order for robots to learn more complex behaviors, recognizing primitive behaviors plays a fundamental role. Research has shown that the recognition of primitive behaviors such as basic gestures enables robots to learn more complex behaviors as combinations of these simple, primitive behaviors. The focus of this study is to investigate the tolerance of neural network models to noisy inputs. We compare and evaluate several neural network architectures including the multilayer perceptron (MLP), time-delay neural network (TDNN), recurrent neural network (RNN) and the Long Short-Term Memory (LSTM). We show that the LSTM is superior to other models in terms of its robustness noisy inputs subjected to Gaussian noise.",
author = "{Dickson Neoh}, {Tze How} and {Mohamed Sahari}, {Khairul Salleh} and Loo, {Chu Kiong}",
year = "2019",
month = "5",
day = "15",
doi = "10.1109/SCIS-ISIS.2018.00142",
language = "English",
series = "Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "859--866",
booktitle = "Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018",
address = "United States",

}

Dickson Neoh, TH, Mohamed Sahari, KS & Loo, CK 2019, Long short-term memory in recognizing behavior sequences on humanoid robot. in Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018., 8716108, Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018, Institute of Electrical and Electronics Engineers Inc., pp. 859-866, Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018, Toyama, Japan, 05/12/18. https://doi.org/10.1109/SCIS-ISIS.2018.00142

Long short-term memory in recognizing behavior sequences on humanoid robot. / Dickson Neoh, Tze How; Mohamed Sahari, Khairul Salleh; Loo, Chu Kiong.

Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018. Institute of Electrical and Electronics Engineers Inc., 2019. p. 859-866 8716108 (Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Long short-term memory in recognizing behavior sequences on humanoid robot.

AU - Dickson Neoh, Tze How

AU - Mohamed Sahari, Khairul Salleh

AU - Loo, Chu Kiong

PY - 2019/5/15

Y1 - 2019/5/15

N2 - In order for robots to learn more complex behaviors, recognizing primitive behaviors plays a fundamental role. Research has shown that the recognition of primitive behaviors such as basic gestures enables robots to learn more complex behaviors as combinations of these simple, primitive behaviors. The focus of this study is to investigate the tolerance of neural network models to noisy inputs. We compare and evaluate several neural network architectures including the multilayer perceptron (MLP), time-delay neural network (TDNN), recurrent neural network (RNN) and the Long Short-Term Memory (LSTM). We show that the LSTM is superior to other models in terms of its robustness noisy inputs subjected to Gaussian noise.

AB - In order for robots to learn more complex behaviors, recognizing primitive behaviors plays a fundamental role. Research has shown that the recognition of primitive behaviors such as basic gestures enables robots to learn more complex behaviors as combinations of these simple, primitive behaviors. The focus of this study is to investigate the tolerance of neural network models to noisy inputs. We compare and evaluate several neural network architectures including the multilayer perceptron (MLP), time-delay neural network (TDNN), recurrent neural network (RNN) and the Long Short-Term Memory (LSTM). We show that the LSTM is superior to other models in terms of its robustness noisy inputs subjected to Gaussian noise.

UR - http://www.scopus.com/inward/record.url?scp=85067129405&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067129405&partnerID=8YFLogxK

U2 - 10.1109/SCIS-ISIS.2018.00142

DO - 10.1109/SCIS-ISIS.2018.00142

M3 - Conference contribution

AN - SCOPUS:85067129405

T3 - Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018

SP - 859

EP - 866

BT - Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Dickson Neoh TH, Mohamed Sahari KS, Loo CK. Long short-term memory in recognizing behavior sequences on humanoid robot. In Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018. Institute of Electrical and Electronics Engineers Inc. 2019. p. 859-866. 8716108. (Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018). https://doi.org/10.1109/SCIS-ISIS.2018.00142