Enhancing Training of Spiking Neural Network with Stochastic Latency

Authors: Srinivas Anumasa, Bhaskar Mukhoty, Velibor Bojkovic, Giulia De Masi, Huan Xiong, Bin Gu

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We provide heuristics for our approach with partial theoretical justification and experimental evidence showing the state-of-the-art performance of our models on datasets such as CIFAR-10, DVS-CIFAR10, CIFAR-100, and DVS-Gesture. Our code is available at https://github.com/srinuvaasu/SLT
Researcher Affiliation Academia 1 Mohamed bin Zayed University of Artificial Intelligence, UAE 2 ARRC, Technology Innovation Institute, UAE 3 Bio Robotics Institute, Sant Anna School of Advanced Studies Pisa, Italy 4 Harbin Institute of Technology, China 5 School of Artificial Intelligence, Jilin University, China
Pseudocode Yes Algorithm 1: SLT: Stochastic Latency Training
Open Source Code Yes Our code is available at https://github.com/srinuvaasu/SLT
Open Datasets Yes Our models on datasets such as CIFAR-10, DVS-CIFAR10, CIFAR-100, and DVS-Gesture. ... CIFAR-10 (Krizhevsky, Hinton et al. 2009) ... DVS-CIFAR-10 (Li et al. 2017) ... DVS-Gesture (Amir et al. 2017)
Dataset Splits No The paper mentions training and test sets (e.g., '5000 train images and 1000 test images' for CIFAR-10), but does not explicitly provide details for a validation split.
Hardware Specification No The paper mentions the use of 'GPUs' for training but does not provide specific hardware models, processors, or detailed specifications.
Software Dependencies No The paper specifies optimizer details like 'Adam' and 'Cosine Ann.' but does not list specific versions of software libraries or programming languages required for replication (e.g., PyTorch version, Python version).
Experiment Setup Yes Table 1: Hyper-parameter settings for comparison lists specific values for 'No. of epochs', 'Mini batch size', 'LIF: β', 'LIF: u0', 'LIF: uth', 'λT ET', 'Optimiser Adam', 'Learning Rate', 'Adam: Betas', 'Rate Scheduler Cosine Ann.'.