Unsupervised Learning in a Ternary SNN Using STDP
This paper proposes a novel implementation of a ternary Spiking Neural Network (SNN) and investigates it using a hierarchical simulation framework. The proposed ternary SNN is trained in an unsupervised manner using the Spike Timing Dependent Plasticity (STDP) learning rule. A ternary neuron is impl...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2024-01-01
|
| Series: | IEEE Journal of the Electron Devices Society |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10437991/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832583219682213888 |
|---|---|
| author | Abhinav Gupta Sneh Saurabh |
| author_facet | Abhinav Gupta Sneh Saurabh |
| author_sort | Abhinav Gupta |
| collection | DOAJ |
| description | This paper proposes a novel implementation of a ternary Spiking Neural Network (SNN) and investigates it using a hierarchical simulation framework. The proposed ternary SNN is trained in an unsupervised manner using the Spike Timing Dependent Plasticity (STDP) learning rule. A ternary neuron is implemented using a Dual-Pocket Tunnel Field effect transistor (DP-TFET). The synapse consists of a Magnetic Tunnel Junction (MTJ) with a Heavy Metal (HM) underlayer, allowing for the adjustment of its conductance by directing a current through the HM layer. Further, we show that a pair of dual-pocket Fully-Depleted Silicon-on-Insulator (FD-SOI) MOSFETs can be utilized to generate a current, which reduces exponentially with increasing duration of firing events between pre- and post-synaptic neurons. This current modulates the synapse’s conductance according to STDP. Furthermore, it is demonstrated that the proposed ternary SNN can be trained to classify digits in the MNIST dataset with an accuracy of 82%, which is better (75%) than that obtained using a binary SNN. Moreover, the runtime required to train the proposed ternary SNN is <inline-formula> <tex-math notation="LaTeX">$8\times $ </tex-math></inline-formula> less than that required for a binary SNN. |
| format | Article |
| id | doaj-art-42eef74052764910930097845619891f |
| institution | Kabale University |
| issn | 2168-6734 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of the Electron Devices Society |
| spelling | doaj-art-42eef74052764910930097845619891f2025-01-29T00:00:11ZengIEEEIEEE Journal of the Electron Devices Society2168-67342024-01-011221122010.1109/JEDS.2024.336619910437991Unsupervised Learning in a Ternary SNN Using STDPAbhinav Gupta0https://orcid.org/0000-0001-8438-4141Sneh Saurabh1https://orcid.org/0000-0002-0587-3391Department of Electronics and Communication Engineering, Indraprastha Institute of Information Technology, Delhi, IndiaDepartment of Electronics and Communication Engineering, Indraprastha Institute of Information Technology, Delhi, IndiaThis paper proposes a novel implementation of a ternary Spiking Neural Network (SNN) and investigates it using a hierarchical simulation framework. The proposed ternary SNN is trained in an unsupervised manner using the Spike Timing Dependent Plasticity (STDP) learning rule. A ternary neuron is implemented using a Dual-Pocket Tunnel Field effect transistor (DP-TFET). The synapse consists of a Magnetic Tunnel Junction (MTJ) with a Heavy Metal (HM) underlayer, allowing for the adjustment of its conductance by directing a current through the HM layer. Further, we show that a pair of dual-pocket Fully-Depleted Silicon-on-Insulator (FD-SOI) MOSFETs can be utilized to generate a current, which reduces exponentially with increasing duration of firing events between pre- and post-synaptic neurons. This current modulates the synapse’s conductance according to STDP. Furthermore, it is demonstrated that the proposed ternary SNN can be trained to classify digits in the MNIST dataset with an accuracy of 82%, which is better (75%) than that obtained using a binary SNN. Moreover, the runtime required to train the proposed ternary SNN is <inline-formula> <tex-math notation="LaTeX">$8\times $ </tex-math></inline-formula> less than that required for a binary SNN.https://ieeexplore.ieee.org/document/10437991/Ternary neuronSTDPternary SNNneuromorphic computingBTBT |
| spellingShingle | Abhinav Gupta Sneh Saurabh Unsupervised Learning in a Ternary SNN Using STDP IEEE Journal of the Electron Devices Society Ternary neuron STDP ternary SNN neuromorphic computing BTBT |
| title | Unsupervised Learning in a Ternary SNN Using STDP |
| title_full | Unsupervised Learning in a Ternary SNN Using STDP |
| title_fullStr | Unsupervised Learning in a Ternary SNN Using STDP |
| title_full_unstemmed | Unsupervised Learning in a Ternary SNN Using STDP |
| title_short | Unsupervised Learning in a Ternary SNN Using STDP |
| title_sort | unsupervised learning in a ternary snn using stdp |
| topic | Ternary neuron STDP ternary SNN neuromorphic computing BTBT |
| url | https://ieeexplore.ieee.org/document/10437991/ |
| work_keys_str_mv | AT abhinavgupta unsupervisedlearninginaternarysnnusingstdp AT snehsaurabh unsupervisedlearninginaternarysnnusingstdp |