三元ディープスパイクニューラルネットワーク

• Item type: SIG Technical Reports(1)
• 公開日: 2022-01-17
• タイトル: 三元ディープスパイクニューラルネットワーク
• タイトル / 言語: en
• タイトル: Ternarizing Deep Spiking Neural Network
• 言語: eng
• キーワード / 主題Scheme: Other
• キーワード / 主題: ニューラルネットワーク
• 資源タイプ / 資源タイプ識別子: http://purl.org/coar/resource_type/c_18gh
• 資源タイプ: technical report
• 著者所属: 奈良先端科学技術大学院大学
• 著者所属: 奈良先端科学技術大学院大学
• 著者所属: 奈良先端科学技術大学院大学
• 著者所属: 奈良先端科学技術大学院大学
• 著者所属: 奈良先端科学技術大学院大学
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者所属(英): en
• 著者所属(英): Nara Institute of Science and Technology
• 著者名: Man, Wu; Yirong, Kan; Vantinh, Nguyen; 張, 任遠; 中島, 康彦
• 著者名(英): Man, Wu; Yirong, Kan; Vantinh, Nguyen; Renyuan, Zhang; Yasuhiko, Nakashima
• 論文抄録 / 内容記述タイプ: Other
• 論文抄録 / 内容記述: The feasibility of ternarizing spiking neural networks (SNNs) is studied in this work toward trading a slight accuracy for significantly reducing computational and memory costs. By leveraging a parametric integrate-and-fire (PIF) neuron with learnable threshold and spike-timing-dependent backpropagation (STDB) learning rule, the ternarized spiking neural networks (TSNNs) enable directly trained with low latency and negligible loss of accuracy. To this end, a paradigm for binary-ternary dot-product operation is realized during the inference; therefore, the TSNNs achieve up to 16x model compression in contrast to the full precision SNNs. Moreover, to mitigate the accuracy gap, an optimized TSNN with a spiking ResNet structure is introduced into TSNN. For proof-of-concept, we evaluate the prototype of proposed TSNN on N-MNIST, CIFAR-10, CIFAR-100, which achieve 98.43%, 89.07%, 65.24% accuracy with 4 timesteps, respectively. On the basis of this prototype, the optimized TSNN improves by 0.84% and 0.51% over CIFAR-10 and CIFAR-100 datasets, respectively.
• 論文抄録(英) / 内容記述タイプ: Other
• 論文抄録(英) / 内容記述: The feasibility of ternarizing spiking neural networks (SNNs) is studied in this work toward trading a slight accuracy for significantly reducing computational and memory costs. By leveraging a parametric integrate-and-fire (PIF) neuron with learnable threshold and spike-timing-dependent backpropagation (STDB) learning rule, the ternarized spiking neural networks (TSNNs) enable directly trained with low latency and negligible loss of accuracy. To this end, a paradigm for binary-ternary dot-product operation is realized during the inference; therefore, the TSNNs achieve up to 16x model compression in contrast to the full precision SNNs. Moreover, to mitigate the accuracy gap, an optimized TSNN with a spiking ResNet structure is introduced into TSNN. For proof-of-concept, we evaluate the prototype of proposed TSNN on N-MNIST, CIFAR-10, CIFAR-100, which achieve 98.43%, 89.07%, 65.24% accuracy with 4 timesteps, respectively. On the basis of this prototype, the optimized TSNN improves by 0.84% and 0.51% over CIFAR-10 and CIFAR-100 datasets, respectively.
• 書誌レコードID / 収録物識別子タイプ: NCID
• 書誌レコードID / 収録物識別子: AN10096105
• 書誌情報: 研究報告システム・アーキテクチャ（ARC）; 巻 2022-ARC-247, 号 14, p. 1-6, 発行日 2022-01-17
• ISSN / 収録物識別子タイプ: ISSN
• ISSN / 収録物識別子: 2188-8574
• Notice: SIG Technical Reports are nonrefereed and hence may later appear in any journals, conferences, symposia, etc.
• 出版者 / 言語: ja
• 出版者: 情報処理学会