Error-backpropagation in temporally encoded networks of spiking neurons

By Sander M. Bohte, Joost N. Kok, Han La Poutré. 2000

1,033 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Cognitive Neuroscience, Electrical and Electronic Engineering

How researchers see this paper

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

In Artificial Intelligence

In Cognitive Neuroscience

In Electrical and Electronic Engineering

Citations in the corpus

Deep learning in neural networks: An overview (2014, methodology): “Many computational models of such spiking neurons have been proposed and analyzed (e.g., Gerstner and van Hemmen, 1992; Zipser et al., 1993; Stemmler, 1996; Tsodyks et al., 1996; Maex and Orban, 1996; Maass, 1996, 1997; Kistler et al., 1997; Amit and Brunel, 1997; Kempter et al., 1999; Song et al., ”
SLAYER: Spike Layer Error Reassignment in Time (2018, methodology): “Event based methods such as SpikeProp [10] and EvSpikeProp [11] have the derivative term defined only around the firing time, whereas [12–14] ignore the temporal effect of spike signal.”
Surrogate Gradient Learning in Spiking Neural Networks: Bringing the Power of Gradient-based optimization to spiking neural networks (2019, background): “For instance, the formulation of SpikeProp [33] required each hidden unit to emit exactly one spike per trial because it is impossible to define firing time for quiescent units.”
The Heidelberg Spiking Data Sets for the Systematic Evaluation of Spiking Neural Networks (2019, background): “Speciﬁcally, spike timing can serve as an extra coding dimension [Bohte et al., 2002, Mostafa, 2018, Comsa et al., 2019].”
A new supervised learning algorithm for multiple spiking neural networks with application in epilepsy and seizure detection (2009, methodology): “In this research, the so-called α-function is used as the spike response function (Bohte et al., 2002; Gerstner & Kistler, 2002):”
HATS: Histograms of Averaged Time Surfaces for Robust Event-Based Object Classification (2018, methodology): “The most commonly used architectures for event-based cameras are Spiking Neural Networks (SNN) [5, 59, 24, 17, 9, 76, 46].”
Introduction to spiking neural networks: Information processing, learning and applications. (2011, methodology): “…classification – classification capabilities of spiking networks trained according to unsupervised learning methods have been tested on the common benchmark datasets, such as, Iris, Wisconsin Breast Cancer or Statlog Landsat dataset (Newman et al. 1998, Bohte et al. 2002a, Belatreche et al. 2003).”
Supervised Learning Based on Temporal Coding in Spiking Neural Networks (2016, methodology): “An approach that bears some similarities to ours is the SpikeProp algorithm [3] that can be used to train spiking networks to produce output spikes at specific times.”
Supervised Learning in Spiking Neural Networks with ReSuMe: Sequence Learning, Classification, and Spike Shifting (2010, methodology): “In the past decade, spiking neural networks have been successfully applied to various classification tasks (Bohte et al., 2002; Ghosh-Dastidar & Adeli, 2007; Gütig & Sompolinsky, 2006; Hopfield & Brody, 2001; Maass et al., 2002; Maass, Natschlaeger, & Markram, 2004; Muresan, 2002).”
A review of learning in biologically plausible spiking neural networks (2019, positive): “Using remote supervised teacher spikes enables ReSuMe to overcome the silent neuron problem existing in the gradient based learning methods such as SpikeProp.”
Deep Learning With Spiking Neurons: Opportunities and Challenges (2018, methodology): “However, SpikeProp has not been applied to problems at the scale of modern deep learning applications, yet, because this method is computationally expensive.”
SuperSpike: Supervised Learning in Multilayer Spiking Neural Networks (2017, background): “SpikeProp (Bohte et al., 2002) sidesteps this problem by defining a differentiable expression on the firing times instead, on which standard gradient descent can be performed.”