A Hierarchical Learning System Incorporating with Supervised, Unsupervised and Reinforcement Learning

Jinglu Hu; Takafumi Sasakawa; Kotaro Hirasawa; Huiru Zheng

doi:10.1007/978-3-540-72383-7_48

A Hierarchical Learning System Incorporating with Supervised, Unsupervised and Reinforcement Learning

Jinglu Hu
, Takafumi Sasakawa
, Kotaro Hirasawa
, Huiru Zheng

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

According to Hebb’s Cell assembly theory, the brain has the capability of function localization. On the other hand, it is suggested that the brain has three different learning paradigms: supervised, unsupervised and reinforcement learning. Inspired by the above knowledge of brain, we present a hierarchical learning system consisting of three parts: supervised learning (SL) part, unsupervised learning (UL) part and reinforcement learning (RL) part. The SL part is a main part learning input-output mapping; the UL part realizes the function localization of learning system by controlling firing strength of neurons in SL part based on input patterns; the RL part optimizes system performance by adjusting parameters in UL part. Simulation results confirm the effectiveness of the proposed hierarchical learning system.

Original language	English
Title of host publication	Advances in Neural Networks – ISNN 2007, Lecture Notes
Place of Publication	Berlin /Heidelberg
Publisher	Springer
Pages	403-412
Number of pages	10
Volume	4491/2
ISBN (Print)	978-3-540-72382-0
DOIs	https://doi.org/10.1007/978-3-540-72383-7_48
Publication status	Published (in print/issue) - Jun 2007
Event	the 4th international symposium on Neural Networks: Advances in Neural Networks - Nanjin China Duration: 1 Jan 2007 → …

Conference

Conference	the 4th international symposium on Neural Networks: Advances in Neural Networks
Period	1/01/07 → …

Keywords

Learning and memory
learning psychology
Learning process
learning theory
mastery learning
systems neuroscience
cognitive processing of abstract concepts

Access to Document

10.1007/978-3-540-72383-7_48

Cite this

@inproceedings{c115372bf6854fbe8dafcc819b72839b,

title = "A Hierarchical Learning System Incorporating with Supervised, Unsupervised and Reinforcement Learning",

abstract = "According to Hebb{\textquoteright}s Cell assembly theory, the brain has the capability of function localization. On the other hand, it is suggested that the brain has three different learning paradigms: supervised, unsupervised and reinforcement learning. Inspired by the above knowledge of brain, we present a hierarchical learning system consisting of three parts: supervised learning (SL) part, unsupervised learning (UL) part and reinforcement learning (RL) part. The SL part is a main part learning input-output mapping; the UL part realizes the function localization of learning system by controlling firing strength of neurons in SL part based on input patterns; the RL part optimizes system performance by adjusting parameters in UL part. Simulation results confirm the effectiveness of the proposed hierarchical learning system.",

keywords = "Learning and memory, learning psychology, Learning process, learning theory, mastery learning, systems neuroscience, cognitive processing of abstract concepts",

author = "Jinglu Hu and Takafumi Sasakawa and Kotaro Hirasawa and Huiru Zheng",

year = "2007",

month = jun,

doi = "10.1007/978-3-540-72383-7\_48",

language = "English",

isbn = "978-3-540-72382-0",

volume = "4491/2",

pages = "403--412",

booktitle = "Advances in Neural Networks – ISNN 2007, Lecture Notes",

publisher = "Springer",

address = "Switzerland",

note = "the 4th international symposium on Neural Networks: Advances in Neural Networks ; Conference date: 01-01-2007",

}

Hu, J, Sasakawa, T, Hirasawa, K & Zheng, H 2007, A Hierarchical Learning System Incorporating with Supervised, Unsupervised and Reinforcement Learning. in Advances in Neural Networks – ISNN 2007, Lecture Notes. vol. 4491/2, Springer, Berlin /Heidelberg, pp. 403-412, the 4th international symposium on Neural Networks: Advances in Neural Networks, 1/01/07. https://doi.org/10.1007/978-3-540-72383-7_48

A Hierarchical Learning System Incorporating with Supervised, Unsupervised and Reinforcement Learning. / Hu, Jinglu; Sasakawa, Takafumi; Hirasawa, Kotaro et al.
Advances in Neural Networks – ISNN 2007, Lecture Notes. Vol. 4491/2 Berlin /Heidelberg: Springer, 2007. p. 403-412.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A Hierarchical Learning System Incorporating with Supervised, Unsupervised and Reinforcement Learning

AU - Hu, Jinglu

AU - Sasakawa, Takafumi

AU - Hirasawa, Kotaro

AU - Zheng, Huiru

PY - 2007/6

Y1 - 2007/6

N2 - According to Hebb’s Cell assembly theory, the brain has the capability of function localization. On the other hand, it is suggested that the brain has three different learning paradigms: supervised, unsupervised and reinforcement learning. Inspired by the above knowledge of brain, we present a hierarchical learning system consisting of three parts: supervised learning (SL) part, unsupervised learning (UL) part and reinforcement learning (RL) part. The SL part is a main part learning input-output mapping; the UL part realizes the function localization of learning system by controlling firing strength of neurons in SL part based on input patterns; the RL part optimizes system performance by adjusting parameters in UL part. Simulation results confirm the effectiveness of the proposed hierarchical learning system.

AB - According to Hebb’s Cell assembly theory, the brain has the capability of function localization. On the other hand, it is suggested that the brain has three different learning paradigms: supervised, unsupervised and reinforcement learning. Inspired by the above knowledge of brain, we present a hierarchical learning system consisting of three parts: supervised learning (SL) part, unsupervised learning (UL) part and reinforcement learning (RL) part. The SL part is a main part learning input-output mapping; the UL part realizes the function localization of learning system by controlling firing strength of neurons in SL part based on input patterns; the RL part optimizes system performance by adjusting parameters in UL part. Simulation results confirm the effectiveness of the proposed hierarchical learning system.

KW - Learning and memory

KW - learning psychology

KW - Learning process

KW - learning theory

KW - mastery learning

KW - systems neuroscience

KW - cognitive processing of abstract concepts

U2 - 10.1007/978-3-540-72383-7_48

DO - 10.1007/978-3-540-72383-7_48

M3 - Conference contribution

SN - 978-3-540-72382-0

VL - 4491/2

SP - 403

EP - 412

BT - Advances in Neural Networks – ISNN 2007, Lecture Notes

PB - Springer

CY - Berlin /Heidelberg

T2 - the 4th international symposium on Neural Networks: Advances in Neural Networks

Y2 - 1 January 2007

ER -

A Hierarchical Learning System Incorporating with Supervised, Unsupervised and Reinforcement Learning

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this