Hierarchical temporal memory introducing time axis in connection segments

Shinichiro Naito, Masafumi Hagiwara

研究成果: Conference contribution

抄録

In this paper, we propose an improved Hierarchical Temporal Memory (HTM) that can consider long-term dependence. HTM is a temporal sequence prediction model imitating the cerebral cortex structure and learning algorithm. This model is composed of cells of a two-dimensional map representing neurons of the brain, and expresses data by a set of cells in an activated state. Further, the data of the next time is predicted from the set of the cells in the predicted state. HTM learns the time series data by updating synapses connecting each cell according to Hebb's rule and keeps the proper relationship of data. In the conventional model, only the connection with the previous data is learned, but in the proposed model the connection with several former data can be learned. The proposed HTM is modified in terms of structure and learning algorithm. In the structure, we introduced a time axis for the segment which is a collection of synapses. About learning algorithm, the connection with several times ago leads the predicted state. As a result of evaluation experiments, it was confirmed that the proposed model can consider longer-term dependency than the conventional model on temporal sequence prediction.

元の言語English
ホスト出版物のタイトルProceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018
出版者Institute of Electrical and Electronics Engineers Inc.
ページ1364-1369
ページ数6
ISBN(電子版)9781538626337
DOI
出版物ステータスPublished - 2019 5 15
イベントJoint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018 - Toyama, Japan
継続期間: 2018 12 52018 12 8

出版物シリーズ

名前Proceedings - 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
Japan
Toyama
期間18/12/518/12/8

Fingerprint

Data storage equipment
Learning Algorithm
Learning algorithms
Synapse
Cell
Model
Cortex
Time Series Data
Prediction Model
Updating
Neuron
Express
Neurons
Time series
Brain
Prediction
Evaluation
Term
Experiment
Experiments

ASJC Scopus subject areas

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

これを引用

Naito, S., & Hagiwara, M. (2019). Hierarchical temporal memory introducing time axis in connection segments. : Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018 (pp. 1364-1369). [8716224] (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.2018.00213

Hierarchical temporal memory introducing time axis in connection segments. / Naito, Shinichiro; Hagiwara, Masafumi.

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. 1364-1369 8716224 (Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018).

研究成果: Conference contribution

Naito, S & Hagiwara, M 2019, Hierarchical temporal memory introducing time axis in connection segments. : Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018., 8716224, 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. 1364-1369, Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018, Toyama, Japan, 18/12/5. https://doi.org/10.1109/SCIS-ISIS.2018.00213
Naito S, Hagiwara M. Hierarchical temporal memory introducing time axis in connection segments. : 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. 1364-1369. 8716224. (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.2018.00213
Naito, Shinichiro ; Hagiwara, Masafumi. / Hierarchical temporal memory introducing time axis in connection segments. 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. 1364-1369 (Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018).
@inproceedings{38a623360c13413ba4fec1f2ceab9b33,
title = "Hierarchical temporal memory introducing time axis in connection segments",
abstract = "In this paper, we propose an improved Hierarchical Temporal Memory (HTM) that can consider long-term dependence. HTM is a temporal sequence prediction model imitating the cerebral cortex structure and learning algorithm. This model is composed of cells of a two-dimensional map representing neurons of the brain, and expresses data by a set of cells in an activated state. Further, the data of the next time is predicted from the set of the cells in the predicted state. HTM learns the time series data by updating synapses connecting each cell according to Hebb's rule and keeps the proper relationship of data. In the conventional model, only the connection with the previous data is learned, but in the proposed model the connection with several former data can be learned. The proposed HTM is modified in terms of structure and learning algorithm. In the structure, we introduced a time axis for the segment which is a collection of synapses. About learning algorithm, the connection with several times ago leads the predicted state. As a result of evaluation experiments, it was confirmed that the proposed model can consider longer-term dependency than the conventional model on temporal sequence prediction.",
keywords = "Cortex learning algorithm, Hierarchical temporal memory, Long-term dependence, Machine learning",
author = "Shinichiro Naito and Masafumi Hagiwara",
year = "2019",
month = "5",
day = "15",
doi = "10.1109/SCIS-ISIS.2018.00213",
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 = "1364--1369",
booktitle = "Proceedings - 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems, SCIS-ISIS 2018",

}

TY - GEN

T1 - Hierarchical temporal memory introducing time axis in connection segments

AU - Naito, Shinichiro

AU - Hagiwara, Masafumi

PY - 2019/5/15

Y1 - 2019/5/15

N2 - In this paper, we propose an improved Hierarchical Temporal Memory (HTM) that can consider long-term dependence. HTM is a temporal sequence prediction model imitating the cerebral cortex structure and learning algorithm. This model is composed of cells of a two-dimensional map representing neurons of the brain, and expresses data by a set of cells in an activated state. Further, the data of the next time is predicted from the set of the cells in the predicted state. HTM learns the time series data by updating synapses connecting each cell according to Hebb's rule and keeps the proper relationship of data. In the conventional model, only the connection with the previous data is learned, but in the proposed model the connection with several former data can be learned. The proposed HTM is modified in terms of structure and learning algorithm. In the structure, we introduced a time axis for the segment which is a collection of synapses. About learning algorithm, the connection with several times ago leads the predicted state. As a result of evaluation experiments, it was confirmed that the proposed model can consider longer-term dependency than the conventional model on temporal sequence prediction.

AB - In this paper, we propose an improved Hierarchical Temporal Memory (HTM) that can consider long-term dependence. HTM is a temporal sequence prediction model imitating the cerebral cortex structure and learning algorithm. This model is composed of cells of a two-dimensional map representing neurons of the brain, and expresses data by a set of cells in an activated state. Further, the data of the next time is predicted from the set of the cells in the predicted state. HTM learns the time series data by updating synapses connecting each cell according to Hebb's rule and keeps the proper relationship of data. In the conventional model, only the connection with the previous data is learned, but in the proposed model the connection with several former data can be learned. The proposed HTM is modified in terms of structure and learning algorithm. In the structure, we introduced a time axis for the segment which is a collection of synapses. About learning algorithm, the connection with several times ago leads the predicted state. As a result of evaluation experiments, it was confirmed that the proposed model can consider longer-term dependency than the conventional model on temporal sequence prediction.

KW - Cortex learning algorithm

KW - Hierarchical temporal memory

KW - Long-term dependence

KW - Machine learning

UR - http://www.scopus.com/inward/record.url?scp=85067129375&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85067129375&partnerID=8YFLogxK

U2 - 10.1109/SCIS-ISIS.2018.00213

DO - 10.1109/SCIS-ISIS.2018.00213

M3 - Conference contribution

AN - SCOPUS:85067129375

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 - 1364

EP - 1369

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 -