IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
Tech. Rep. Archives
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2019-10-03 11:15
[Poster Presentation] Deep Reinforcement Learning-Based Channel Allocation for Wireless LANs -- Investigation of sampling method of replay buffer --
Kota Nakashima, Shotaro Kamiya, Ohtsu Kazuki, Koji Yamamoto, Takayuki Nishio, Masahiro Morikura (Kyoto Univ.)
Abstract (in Japanese) (See Japanese page) 
(in English) We have proposed the deep reinforcement learning-based channel allocation approach for wireless area local networks to improve the throughput. This report proposes a selective replay buffer storing method for a deep reinforcement learning-based channel allocation approach to avoid the over fitting. When the agent selects actions based on a policy that mostly has the optimal response, a fixed action tends to be selected in a state. The unbalanced learning data can proceed learning only for some certain states and reduce the performance for the other states, which is called the over-fitting. We introduce a selective replay buffer storing method that prevents the unbalance of learning data and the over-fitting, by reducing the storing number of the same data. The simulation results show that the proposal method can prevent the over-fitting and improve the learning performance.
Keyword (in Japanese) (See Japanese page) 
(in English) deep reinforcement learning / over-fitting / replay buffer / / / / /  
Reference Info. IEICE Tech. Rep.
Paper #  
Date of Issue  
ISSN  
Download PDF

Conference Information
Committee MIKA  
Conference Date 2019-10-02 - 2019-10-04 
Place (in Japanese) (See Japanese page) 
Place (in English) Hokkaido Univ. 
Topics (in Japanese) (See Japanese page) 
Topics (in English) Wireless Communication System, etc. 
Paper Information
Registration To MIKA 
Conference Code 2019-10-MIKA 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Deep Reinforcement Learning-Based Channel Allocation for Wireless LANs 
Sub Title (in English) Investigation of sampling method of replay buffer 
Keyword(1) deep reinforcement learning  
Keyword(2) over-fitting  
Keyword(3) replay buffer  
Keyword(4)  
Keyword(5)  
Keyword(6)  
Keyword(7)  
Keyword(8)  
1st Author's Name Kota Nakashima  
1st Author's Affiliation Kyoto University (Kyoto Univ.)
2nd Author's Name Shotaro Kamiya  
2nd Author's Affiliation Kyoto University (Kyoto Univ.)
3rd Author's Name Ohtsu Kazuki  
3rd Author's Affiliation Kyoto University (Kyoto Univ.)
4th Author's Name Koji Yamamoto  
4th Author's Affiliation Kyoto University (Kyoto Univ.)
5th Author's Name Takayuki Nishio  
5th Author's Affiliation Kyoto University (Kyoto Univ.)
6th Author's Name Masahiro Morikura  
6th Author's Affiliation Kyoto University (Kyoto Univ.)
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
Speaker Author-1 
Date Time 2019-10-03 11:15:00 
Presentation Time 50 minutes 
Registration for MIKA 
Paper #  
Volume (vol) vol. 
Number (no)  
Page  
#Pages  
Date of Issue  


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan