Balancing excitation and inhibition of spike neuron using deep Q network (DQN)
Deep reinforcement learning which involved reinforcement learning with artificial neural networks allows an agent to take the best possible actions in a virtual environment to achieve goals. Spike neuron has a non-differentiable spike generation function that caused SNN training faced difficulty. In...
Saved in:
Main Authors: | , , , , |
---|---|
Format: | Conference or Workshop Item |
Language: | English |
Published: |
IOP Publishing
2021
|
Subjects: | |
Online Access: | http://umpir.ump.edu.my/id/eprint/31934/1/Balancing%20excitation%20and%20inhibition%20of%20spike%20neuron.pdf http://umpir.ump.edu.my/id/eprint/31934/ https://doi.org/10.1088/1742-6596/1755/1/012004 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.ump.umpir.31934 |
---|---|
record_format |
eprints |
spelling |
my.ump.umpir.319342021-09-07T04:59:02Z http://umpir.ump.edu.my/id/eprint/31934/ Balancing excitation and inhibition of spike neuron using deep Q network (DQN) Tan, Szi Hui Ishak, Mohamad Khairi Packeer Mohamed, Mohamed Fauzi Mohd Fadzil, Lokman Ahmad Afif, Ahmarofi QA76 Computer software TK Electrical engineering. Electronics Nuclear engineering Deep reinforcement learning which involved reinforcement learning with artificial neural networks allows an agent to take the best possible actions in a virtual environment to achieve goals. Spike neuron has a non-differentiable spike generation function that caused SNN training faced difficulty. In order to overcome the difficulty, Deep Q Network (DQN) is proposed to act as an agent to interact with a custom environment. A spike neuron is modelled by using NEST simulator. Rewards are given to the agent for every action taken. The model is trained and tested to validate the performance of the trained model in order to attain balance the firing rate of excitatory and inhibitory population of spike neuron. Training result showed the agent able to handle the environment. The trained model capable to balance the excitation and inhibition of the spike neuron as the actual output neuron rate is close to or same with the target neuron firing rate. The average percentage error of rate of difference between output and target neuron rate for 5 episodes achieved 0.80%. IOP Publishing 2021-03-01 Conference or Workshop Item PeerReviewed pdf en cc_by http://umpir.ump.edu.my/id/eprint/31934/1/Balancing%20excitation%20and%20inhibition%20of%20spike%20neuron.pdf Tan, Szi Hui and Ishak, Mohamad Khairi and Packeer Mohamed, Mohamed Fauzi and Mohd Fadzil, Lokman and Ahmad Afif, Ahmarofi (2021) Balancing excitation and inhibition of spike neuron using deep Q network (DQN). In: Journal of Physics: Conference Series, 5th International Conference on Electronic Design (ICED 2020), 19 August 2020 , Perlis (Virtual Mode). pp. 1-16., 1755 (012004). ISSN 1742-6588 (print); 1742-6596 (online) https://doi.org/10.1088/1742-6596/1755/1/012004 |
institution |
Universiti Malaysia Pahang |
building |
UMP Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Malaysia Pahang |
content_source |
UMP Institutional Repository |
url_provider |
http://umpir.ump.edu.my/ |
language |
English |
topic |
QA76 Computer software TK Electrical engineering. Electronics Nuclear engineering |
spellingShingle |
QA76 Computer software TK Electrical engineering. Electronics Nuclear engineering Tan, Szi Hui Ishak, Mohamad Khairi Packeer Mohamed, Mohamed Fauzi Mohd Fadzil, Lokman Ahmad Afif, Ahmarofi Balancing excitation and inhibition of spike neuron using deep Q network (DQN) |
description |
Deep reinforcement learning which involved reinforcement learning with artificial neural networks allows an agent to take the best possible actions in a virtual environment to achieve goals. Spike neuron has a non-differentiable spike generation function that caused SNN training faced difficulty. In order to overcome the difficulty, Deep Q Network (DQN) is proposed to act as an agent to interact with a custom environment. A spike neuron is modelled by using NEST simulator. Rewards are given to the agent for every action taken. The model is trained and tested to validate the performance of the trained model in order to attain balance the firing rate of excitatory and inhibitory population of spike neuron. Training result showed the agent able to handle the environment. The trained model capable to balance the excitation and inhibition of the spike neuron as the actual output neuron rate is close to or same with the target neuron firing rate. The average percentage error of rate of difference between output and target neuron rate for 5 episodes achieved 0.80%. |
format |
Conference or Workshop Item |
author |
Tan, Szi Hui Ishak, Mohamad Khairi Packeer Mohamed, Mohamed Fauzi Mohd Fadzil, Lokman Ahmad Afif, Ahmarofi |
author_facet |
Tan, Szi Hui Ishak, Mohamad Khairi Packeer Mohamed, Mohamed Fauzi Mohd Fadzil, Lokman Ahmad Afif, Ahmarofi |
author_sort |
Tan, Szi Hui |
title |
Balancing excitation and inhibition of spike neuron using deep Q network (DQN) |
title_short |
Balancing excitation and inhibition of spike neuron using deep Q network (DQN) |
title_full |
Balancing excitation and inhibition of spike neuron using deep Q network (DQN) |
title_fullStr |
Balancing excitation and inhibition of spike neuron using deep Q network (DQN) |
title_full_unstemmed |
Balancing excitation and inhibition of spike neuron using deep Q network (DQN) |
title_sort |
balancing excitation and inhibition of spike neuron using deep q network (dqn) |
publisher |
IOP Publishing |
publishDate |
2021 |
url |
http://umpir.ump.edu.my/id/eprint/31934/1/Balancing%20excitation%20and%20inhibition%20of%20spike%20neuron.pdf http://umpir.ump.edu.my/id/eprint/31934/ https://doi.org/10.1088/1742-6596/1755/1/012004 |
_version_ |
1710675696630628352 |
score |
13.160551 |