Please use this identifier to cite or link to this item:
https://doi.org/10.1145/3477145.3477157
Title: | Connection Pruning for Deep Spiking Neural Networks with On-Chip Learning | Authors: | Nguyen, TNN Veeravalli, B Fong, X |
Issue Date: | 27-Jul-2021 | Publisher: | ACM | Citation: | Nguyen, TNN, Veeravalli, B, Fong, X (2021-07-27). Connection Pruning for Deep Spiking Neural Networks with On-Chip Learning. ICONS 2021: International Conference on Neuromorphic Systems 2021. ScholarBank@NUS Repository. https://doi.org/10.1145/3477145.3477157 | Abstract: | Long training time hinders the potential of the deep, large-scale Spiking Neural Network (SNN) with the on-chip learning capability to be realized on the embedded systems hardware. Our work proposes a novel connection pruning approach that can be applied during the on-chip Spike Timing Dependent Plasticity (STDP)-based learning to optimize the learning time and the network connectivity of the deep SNN. We applied our approach to a deep SNN with the Time To First Spike (TTFS) coding and has successfully achieved 2.1x speed-up and 64% energy savings in the on-chip learning and reduced the network connectivity by 92.83%, without incurring any accuracy loss. Moreover, the connectivity reduction results in 2.83x speed-up and 78.24% energy savings in the inference. Evaluation of our proposed approach on the Field Programmable Gate Array (FPGA) platform revealed 0.56% power overhead was needed to implement the pruning algorithm. | Source Title: | ICONS 2021: International Conference on Neuromorphic Systems 2021 | URI: | https://scholarbank.nus.edu.sg/handle/10635/245763 | ISBN: | 9781450386913 | DOI: | 10.1145/3477145.3477157 |
Appears in Collections: | Staff Publications Elements |
Show full item record
Files in This Item:
File | Description | Size | Format | Access Settings | Version | |
---|---|---|---|---|---|---|
3477145.3477157.pdf | Published version | 920.01 kB | Adobe PDF | CLOSED | None |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.