Matches in SemOpenAlex for { <https://semopenalex.org/work/W3084655686> ?p ?o ?g. }
- W3084655686 abstract "We derive the backpropagation algorithm for spiking neural networks composed of leaky integrate-and-fire neurons operating in continuous time. This algorithm, EventProp, computes the exact gradient of an arbitrary loss function of spike times and membrane potentials by backpropagating errors in time. For the first time, by leveraging methods from optimal control theory, we are able to backpropagate errors through spike discontinuities without approximations or smoothing operations. As errors are backpropagated in an event-based manner (at spike times), EventProp requires storing state variables only at these times, providing favorable memory requirements. EventProp can be applied to spiking networks with arbitrary connectivity, including recurrent, convolutional and deep feed-forward architectures. While we consider the leaky integrate-and-fire neuron model in this work, our methodology to derive the gradient can be applied to other spiking neuron models. We demonstrate learning using gradients computed via EventProp in a deep spiking network using an event-based simulator and a non-linearly separable dataset encoded using spike time latencies. Our work supports the rigorous study of gradient-based methods to train spiking neural networks while providing insights toward the development of learning algorithms in neuromorphic hardware." @default.
- W3084655686 created "2020-09-21" @default.
- W3084655686 creator A5050178239 @default.
- W3084655686 creator A5077874923 @default.
- W3084655686 date "2020-09-17" @default.
- W3084655686 modified "2023-09-27" @default.
- W3084655686 title "EventProp: Backpropagation for Exact Gradients in Spiking Neural Networks" @default.
- W3084655686 cites W1522301498 @default.
- W3084655686 cites W1542343170 @default.
- W3084655686 cites W1587799944 @default.
- W3084655686 cites W1597009674 @default.
- W3084655686 cites W1965678517 @default.
- W3084655686 cites W1970109917 @default.
- W3084655686 cites W1979489469 @default.
- W3084655686 cites W2008658242 @default.
- W3084655686 cites W2081209729 @default.
- W3084655686 cites W2094731495 @default.
- W3084655686 cites W2100324194 @default.
- W3084655686 cites W2101234009 @default.
- W3084655686 cites W2112796928 @default.
- W3084655686 cites W2115831804 @default.
- W3084655686 cites W2138913040 @default.
- W3084655686 cites W2147792971 @default.
- W3084655686 cites W2149530067 @default.
- W3084655686 cites W2159951683 @default.
- W3084655686 cites W2165312143 @default.
- W3084655686 cites W2290982066 @default.
- W3084655686 cites W2314470091 @default.
- W3084655686 cites W2316171798 @default.
- W3084655686 cites W2317926736 @default.
- W3084655686 cites W2513853720 @default.
- W3084655686 cites W2749476078 @default.
- W3084655686 cites W2769049661 @default.
- W3084655686 cites W2783525259 @default.
- W3084655686 cites W2798878556 @default.
- W3084655686 cites W2808746463 @default.
- W3084655686 cites W2891530223 @default.
- W3084655686 cites W2898323475 @default.
- W3084655686 cites W2900163261 @default.
- W3084655686 cites W2907149254 @default.
- W3084655686 cites W2938310555 @default.
- W3084655686 cites W2944119451 @default.
- W3084655686 cites W2946725528 @default.
- W3084655686 cites W2948318 @default.
- W3084655686 cites W2952754147 @default.
- W3084655686 cites W2962908174 @default.
- W3084655686 cites W2963335874 @default.
- W3084655686 cites W2963834742 @default.
- W3084655686 cites W2971713705 @default.
- W3084655686 cites W2984844508 @default.
- W3084655686 cites W2990793844 @default.
- W3084655686 cites W3015205410 @default.
- W3084655686 cites W3089433719 @default.
- W3084655686 cites W3102087395 @default.
- W3084655686 cites W3104409553 @default.
- W3084655686 cites W3124478039 @default.
- W3084655686 cites W3144553733 @default.
- W3084655686 cites W3191749578 @default.
- W3084655686 cites W345956458 @default.
- W3084655686 cites W59997565 @default.
- W3084655686 cites W2006370340 @default.
- W3084655686 hasPublicationYear "2020" @default.
- W3084655686 type Work @default.
- W3084655686 sameAs 3084655686 @default.
- W3084655686 citedByCount "4" @default.
- W3084655686 countsByYear W30846556862020 @default.
- W3084655686 countsByYear W30846556862021 @default.
- W3084655686 crossrefType "posted-content" @default.
- W3084655686 hasAuthorship W3084655686A5050178239 @default.
- W3084655686 hasAuthorship W3084655686A5077874923 @default.
- W3084655686 hasConcept C108583219 @default.
- W3084655686 hasConcept C11413529 @default.
- W3084655686 hasConcept C115903868 @default.
- W3084655686 hasConcept C11731999 @default.
- W3084655686 hasConcept C121332964 @default.
- W3084655686 hasConcept C134306372 @default.
- W3084655686 hasConcept C151927369 @default.
- W3084655686 hasConcept C154945302 @default.
- W3084655686 hasConcept C155032097 @default.
- W3084655686 hasConcept C15627037 @default.
- W3084655686 hasConcept C2779662365 @default.
- W3084655686 hasConcept C2781390188 @default.
- W3084655686 hasConcept C2909946758 @default.
- W3084655686 hasConcept C33923547 @default.
- W3084655686 hasConcept C38365724 @default.
- W3084655686 hasConcept C41008148 @default.
- W3084655686 hasConcept C50644808 @default.
- W3084655686 hasConcept C62520636 @default.
- W3084655686 hasConcept C81363708 @default.
- W3084655686 hasConceptScore W3084655686C108583219 @default.
- W3084655686 hasConceptScore W3084655686C11413529 @default.
- W3084655686 hasConceptScore W3084655686C115903868 @default.
- W3084655686 hasConceptScore W3084655686C11731999 @default.
- W3084655686 hasConceptScore W3084655686C121332964 @default.
- W3084655686 hasConceptScore W3084655686C134306372 @default.
- W3084655686 hasConceptScore W3084655686C151927369 @default.
- W3084655686 hasConceptScore W3084655686C154945302 @default.
- W3084655686 hasConceptScore W3084655686C155032097 @default.
- W3084655686 hasConceptScore W3084655686C15627037 @default.
- W3084655686 hasConceptScore W3084655686C2779662365 @default.