Spikemax: Spike-based Loss Methods for Classification

被引:3
|
作者
Shrestha, Sumit Bam [1 ]
Zhu, Longwei [1 ]
Sun, Pengfei [1 ,2 ]
机构
[1] ASTAR, Inst Infocomm Res, Singapore, Singapore
[2] Univ Ghent, Ghent, Belgium
关键词
NEURAL-NETWORKS; MODEL;
D O I
10.1109/IJCNN55064.2022.9892379
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Spiking Neural Networks (SNNs) are a promising research paradigm for low power edge-based computing. Recent works in SNN backpropagation has enabled training of SNNs for practical tasks. However, since spikes are binary events in time, standard loss formulations are not directly compatible with spike output. As a result, current works are limited to using mean-squared loss of spike count. In this paper, we formulate the output probability interpretation from the spike count measure and introduce spike-based negative log-likelihood measure which are more suited for classification tasks especially in terms of the energy efficiency and inference latency. We compare our loss measures with other existing alternatives and evaluate using classification performances on three neuromorphic benchmark datasets: NMNIST, DVS Gesture and N-TIDIGITS18. In addition, we demonstrate state of the art performances on these datasets, achieving faster inference speed and less energy consumption.
引用
收藏
页数:7
相关论文
共 50 条
  • [21] A Neuromorphic VLSI Circuit for Spike-Based Random Sampling
    Chien, Chen-Han
    Liu, Shih-Chii
    Steimer, Andreas
    IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTING, 2018, 6 (01) : 135 - 144
  • [22] Learning spike-based correlations and conditional probabilities in silicon
    Shon, AP
    Hsu, D
    Diorio, C
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 14, VOLS 1 AND 2, 2002, 14 : 1123 - 1130
  • [23] Easy and efficient spike-based Machine Learning with mlGeNN
    Knight, James C.
    Nowotny, Thomas
    PROCEEDINGS OF THE 2023 ANNUAL NEURO-INSPIRED COMPUTATIONAL ELEMENTS CONFERENCE, NICE 2023, 2023, : 115 - 120
  • [24] An AER to CAN Bridge for Spike-Based Robot Control
    Dominguez-Morales, M.
    Jimenez-Fernandez, A.
    Paz, R.
    Linares-Barranco, A.
    Cascado, D.
    Coronado, J. L.
    Munoz, J. L.
    Jimenez, G.
    ADVANCES IN COMPUTATIONAL INTELLIGENCE, IWANN 2011, PT I, 2011, 6691 : 124 - 132
  • [25] FPGA implementation of a spike-based sound localization system
    Ponca, M
    Schauer, C
    ARTIFICIAL NEURAL NETS AND GENETIC ALGORITHMS, 2001, : 212 - 215
  • [26] Gated parametric neuron for spike-based audio recognition
    Wang, Haoran
    Zhang, Herui
    Li, Siyang
    Wu, Dongrui
    NEUROCOMPUTING, 2024, 609
  • [27] Towards spike-based machine intelligence with neuromorphic computing
    Roy, Kaushik
    Jaiswal, Akhilesh
    Panda, Priyadarshini
    NATURE, 2019, 575 (7784) : 607 - 617
  • [28] Spike-based compared to rate-based Hebbian learning
    Kempter, R
    Gerstner, W
    van Hemmen, JL
    ADVANCES IN NEURAL INFORMATION PROCESSING SYSTEMS 11, 1999, 11 : 125 - 131
  • [29] Benchmarking Spike-Based Visual Recognition: A Dataset and Evaluation
    Liu, Qian
    Pineda-Garcia, Garibaldi
    Stromatias, Evangelos
    Serrano-Gotarredona, Teresa
    Furber, Steve B.
    FRONTIERS IN NEUROSCIENCE, 2016, 10
  • [30] Live Demonstration: Computing Spike-based Convolutions on GPUs
    Nageswaran, Jayram Moorkanikara
    Dutt, Nikil
    Wang, Yingxue
    Delbrueck, Tobi
    ISCAS: 2009 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS, VOLS 1-5, 2009, : 1921 - 1921