Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Supervised learning of spatial features with STDP and homeostasis using Spiking Neural Networks on SpiNNaker

Sergio Davies, Andrew Gait, Andrew Rowley, Alessandro Di Nuovo

TL;DR

A new method to perform supervised learning on SNNs, using Spike Timing Dependent Plasticity (STDP) and homeostasis, aiming at training the network to identify spatial patterns, may be applied to pattern recognition in static images or traffic analysis in computer networks.

Abstract

Artificial Neural Networks (ANN) have gained significant popularity thanks to their ability to learn using the well-known backpropagation algorithm. Conversely, Spiking Neural Networks (SNNs), despite having broader capabilities than ANNs, have always posed challenges in the training phase. This paper shows a new method to perform supervised learning on SNNs, using Spike Timing Dependent Plasticity (STDP) and homeostasis, aiming at training the network to identify spatial patterns. Spatial patterns refer to spike patterns without a time component, where all spike events occur simultaneously. The method is tested using the SpiNNaker digital architecture. A SNN is trained to recognise one or multiple patterns and performance metrics are extracted to measure the performance of the network. Some considerations are drawn from the results showing that, in the case of a single trained pattern, the network behaves as the ideal detector, with 100% accuracy in detecting the trained pattern. However, as the number of trained patterns on a single network increases, the accuracy of identification is linked to the similarities between these patterns. This method of training an SNN to detect spatial patterns may be applied to pattern recognition in static images or traffic analysis in computer networks, where each network packet represents a spatial pattern. It will be stipulated that the homeostatic factor may enable the network to detect patterns with some degree of similarity, rather than only perfectly matching patterns.The principles outlined in this article serve as the fundamental building blocks for more complex systems that utilise both spatial and temporal patterns by converting specific features of input signals into spikes.One example of such a system is a computer network packet classifier, tasked with real-time identification of packet streams based on features within the packet content

Supervised learning of spatial features with STDP and homeostasis using Spiking Neural Networks on SpiNNaker

TL;DR

A new method to perform supervised learning on SNNs, using Spike Timing Dependent Plasticity (STDP) and homeostasis, aiming at training the network to identify spatial patterns, may be applied to pattern recognition in static images or traffic analysis in computer networks.

Abstract

Artificial Neural Networks (ANN) have gained significant popularity thanks to their ability to learn using the well-known backpropagation algorithm. Conversely, Spiking Neural Networks (SNNs), despite having broader capabilities than ANNs, have always posed challenges in the training phase. This paper shows a new method to perform supervised learning on SNNs, using Spike Timing Dependent Plasticity (STDP) and homeostasis, aiming at training the network to identify spatial patterns. Spatial patterns refer to spike patterns without a time component, where all spike events occur simultaneously. The method is tested using the SpiNNaker digital architecture. A SNN is trained to recognise one or multiple patterns and performance metrics are extracted to measure the performance of the network. Some considerations are drawn from the results showing that, in the case of a single trained pattern, the network behaves as the ideal detector, with 100% accuracy in detecting the trained pattern. However, as the number of trained patterns on a single network increases, the accuracy of identification is linked to the similarities between these patterns. This method of training an SNN to detect spatial patterns may be applied to pattern recognition in static images or traffic analysis in computer networks, where each network packet represents a spatial pattern. It will be stipulated that the homeostatic factor may enable the network to detect patterns with some degree of similarity, rather than only perfectly matching patterns.The principles outlined in this article serve as the fundamental building blocks for more complex systems that utilise both spatial and temporal patterns by converting specific features of input signals into spikes.One example of such a system is a computer network packet classifier, tasked with real-time identification of packet streams based on features within the packet content
Paper Structure (10 sections, 3 equations, 6 figures, 10 tables)

This paper contains 10 sections, 3 equations, 6 figures, 10 tables.

Table of Contents

  1. Introduction
  2. Methodology
  3. STDP on SpiNNaker
  4. Training phase
  5. Test and Validation phase
  6. Results
  7. Single pattern training
  8. Dual pattern training
  9. Multiple pattern training
  10. Conclusions

Figures (6)

  • Figure 1: An example of how weight change (dW) is calculated based on the time difference between pre- and post-synaptic spikes (dt). In green on the left is the Long Term Potentiation (LTP) generated by a pre-synaptic and post-synaptic spike sequence. In blue on the right is the Long Term Depression (LTD) generated by a post-synaptic and pre-synaptic spike sequence.
  • Figure 2: The sequence of spikes in the network used for training.
  • Figure 3: The network used for training. In blue the STDP-enabled synapses.
  • Figure 4: The precise timing of the spikes for potentiation and depression. The timing for both Long Term Potentiation (LTP) and Long Term Depression (LTD) is considered on the timestep immediately following the outgoing spike value, so the values concerned here are 5 ms when the "1" is potentiated and the "0" is depressed, and 25 ms when the "0" is potentiated and the "1" is depressed.
  • Figure 5: The network used for testing.
  • ...and 1 more figures