STDP-based spiking deep neural networks for object recognition

By S. R. Kheradpisheh, Mohammad Ganjtabesh, Simon J. Thorpe, Timothée Masquelier. Neural Networks, 2016

458 citations recorded by the source corpus.

Previous studies have shown that spike-timing-dependent plasticity (STDP) can be used in spiking neural networks (SNN) to extract visual features of low or intermediate complexity in an unsupervised manner. These studies, however, used relatively shallow architectures, and only one layer was trainable. Another line of research has demonstrated - using rate-based neural networks trained with back-propagation - that having many layers increases the recognition robustness, an approach known as deep learning. We thus designed a deep SNN, comprising several convolutional (trainable with STDP) and pooling layers. We used a temporal coding scheme where the most strongly activated neurons fire first, and less activated neurons fire later or not at all. The network was exposed to natural images. Thanks to STDP, neurons progressively learned features corresponding to prototypical patterns that were both salient and frequent. Only a few tens of examples per category were required and no label was needed. After learning, the complexity of the extracted features increased along the hierarchy, from edge detectors in the first layer to object prototypes in the last layer. Coding was very sparse, with only a few thousands spikes per image, and in some cases the object category could be reasonably well inferred from the activity of a single higher-order neuron. More generally, the activity of a few hundreds of such neurons contained robust category information, as demonstrated using a classifier on Caltech 101, ETH-80, and MNIST databases. We also demonstrate the superiority of STDP over other unsupervised techniques such as random crops (HMAX) or auto-encoders. Taken together, our results suggest that the combination of STDP with latency coding may be a key to understanding the way that the primate visual system learns, its remarkable processing speed and its low energy consumption. These mechanisms are also interesting for artificial vision systems, particularly for hardware solutions.

Subfields: Cognitive Neuroscience, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/19751e0f81a103658bbac2506f5d5c8e06a1c06a

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Electrical and Electronic Engineering

Citations in the corpus

Deep Learning With Spiking Neurons: Opportunities and Challenges (2018, methodology): “The BrainScaleS system (Schemmel et al., 2010) and its successor (Aamir et al., 2018) is an especially promising candidate for on-chip learning due to its acceleration of up to a factor of 10000 compared to biological real time, but so far STDP is only shown for small networks on a prototype chip (P”
Deep Learning in Spiking Neural Networks (2018, methodology): “convolutional/pooling layers trained by STDP [170].”
Spiking Neural Networks and Their Applications: A Review (2022, background): “Recently, some remarkable studies have applied SNNs for object detection tasks [72–74], showing comparative results with DNNs while requiring much less energy for the computations.”
A review of learning in biologically plausible spiking neural networks (2019, unknown): “The final pooling layer in the SCNN is a global max pooling layer which pool on overall neuronal maps of the last convolutional layer.”