Improvement of pattern recognition in spiking neural networks by modifying threshold parameter and using image inversion

Aghabarar, Hedyeh; Kiani, Kourosh; Keshavarzi, Parviz

doi:10.1007/s11042-023-16344-3

Cited by 4 publications

(2 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the simulations conducted on the proposed model, the parameter values used for the neuronal model features are listed in Table 1. For some of these parameters, the effect of their different values is examined on the recognition accuracy of the model to help us gain a better understanding of the model's functionality and the impact of each parameter on its performance [20], which will be explained in detail later.…”

Section: -Results and Discussionmentioning

confidence: 99%

“…This paper improved the pattern recognition accuracy in this model by applying the eligibility traces method on reinforcement learning and also adjusting several intrinsic neuron parameters, i.e. hyperparameters, which one of them is considered in [20]. This includes changing the increment value of the threshold voltage of AdLIF neurons in the network and changing the time constant and the threshold voltage of simple and adaptive LIF neurons of the network concerning their rest and reset voltage.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Reinforcement Learning in Deep Spiking Neural Networks with Eligibility Traces and Modifying the Threshold Parameter

Aghabarar,

Keshavarzi,

Kiani

2024

Preprint

Self Cite

View full text Add to dashboard Cite

Desirable features of Spiking Neural Networks (SNNs) such as biological plausibility, event-based information processing, and energy efficiency have led to their widespread application in various machine learning domains in recent years. The reinforcement learning, inspired by the human visual system, has also gained popularity recently. This paper proposed reinforcement learning in a deep SNN which presents the idea of using eligibility traces to apply reward signals. We also focus on a locally connected SNN that uses a reinforcement learning based on Spike-Timing-Dependent Plasticity (STDP) called R-STDP for pattern learning. Locally connected networks, exhibit a closer resemblance to the biological visual system by extracting key topological features of the image. By introducing this idea and modifying several intrinsic parameters of spiking neurons (adjusting the voltage threshold, as well as the membrane time constant), the model achieves an 87.84% accuracy on MNIST dataset, showing significant improvement compared to the previous similar model.

show abstract

Section: -Results and Discussionmentioning

confidence: 99%

mentioning

confidence: 99%