Matches in SemOpenAlex for { <https://semopenalex.org/work/W4226474429> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W4226474429 endingPage "2044" @default.
- W4226474429 startingPage "2034" @default.
- W4226474429 abstract "Spiking neural networks (SNNs) capture some of the efficiency of biological brains for inference and learning via the dynamic, online, and event-driven processing of binary time series. Most existing learning algorithms for SNNs are based on deterministic neuronal models, such as leaky integrate-and-fire, and rely on heuristic approximations of backpropagation through time that enforces constraints such as locality. In contrast, probabilistic SNN models can be trained directly via principled online, local, and update rules that have proven to be particularly effective for resource-constrained systems. This article investigates another advantage of probabilistic SNNs, namely, their capacity to generate independent outputs when queried over the same input. It is shown that the multiple generated output samples can be used during inference to robustify decisions and to quantify uncertainty-a feature that deterministic SNN models cannot provide. Furthermore, they can be leveraged for training in order to obtain more accurate statistical estimates of the log-loss training criterion and its gradient. Specifically, this article introduces an online learning rule based on generalized expectation-maximization (GEM) that follows a three-factor form with global learning signals and is referred to as GEM-SNN. Experimental results on structured output memorization and classification on a standard neuromorphic dataset demonstrate significant improvements in terms of log-likelihood, accuracy, and calibration when increasing the number of samples used for inference and training." @default.
- W4226474429 created "2022-05-05" @default.
- W4226474429 creator A5017736224 @default.
- W4226474429 creator A5056471397 @default.
- W4226474429 date "2022-05-01" @default.
- W4226474429 modified "2023-10-02" @default.
- W4226474429 title "Multisample Online Learning for Probabilistic Spiking Neural Networks" @default.
- W4226474429 cites W2069519142 @default.
- W4226474429 cites W2069845225 @default.
- W4226474429 cites W2093002383 @default.
- W4226474429 cites W2111562869 @default.
- W4226474429 cites W2138913040 @default.
- W4226474429 cites W2165396124 @default.
- W4226474429 cites W2207801947 @default.
- W4226474429 cites W2271476098 @default.
- W4226474429 cites W2567948266 @default.
- W4226474429 cites W2569813014 @default.
- W4226474429 cites W2783525259 @default.
- W4226474429 cites W2801490189 @default.
- W4226474429 cites W2963437309 @default.
- W4226474429 cites W2963569875 @default.
- W4226474429 cites W2963809228 @default.
- W4226474429 cites W2969335882 @default.
- W4226474429 cites W2984844508 @default.
- W4226474429 cites W3025773901 @default.
- W4226474429 cites W3100658279 @default.
- W4226474429 cites W3101210313 @default.
- W4226474429 cites W3102087395 @default.
- W4226474429 cites W3118913981 @default.
- W4226474429 cites W3119354933 @default.
- W4226474429 cites W3119431736 @default.
- W4226474429 cites W3159341886 @default.
- W4226474429 cites W3160817857 @default.
- W4226474429 cites W4205197926 @default.
- W4226474429 cites W4234572373 @default.
- W4226474429 doi "https://doi.org/10.1109/tnnls.2022.3144296" @default.
- W4226474429 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/35089867" @default.
- W4226474429 hasPublicationYear "2022" @default.
- W4226474429 type Work @default.
- W4226474429 citedByCount "3" @default.
- W4226474429 countsByYear W42264744292022 @default.
- W4226474429 countsByYear W42264744292023 @default.
- W4226474429 crossrefType "journal-article" @default.
- W4226474429 hasAuthorship W4226474429A5017736224 @default.
- W4226474429 hasAuthorship W4226474429A5056471397 @default.
- W4226474429 hasBestOaLocation W42264744292 @default.
- W4226474429 hasConcept C11731999 @default.
- W4226474429 hasConcept C119857082 @default.
- W4226474429 hasConcept C154945302 @default.
- W4226474429 hasConcept C173801870 @default.
- W4226474429 hasConcept C2776214188 @default.
- W4226474429 hasConcept C2779127903 @default.
- W4226474429 hasConcept C41008148 @default.
- W4226474429 hasConcept C49937458 @default.
- W4226474429 hasConcept C50644808 @default.
- W4226474429 hasConceptScore W4226474429C11731999 @default.
- W4226474429 hasConceptScore W4226474429C119857082 @default.
- W4226474429 hasConceptScore W4226474429C154945302 @default.
- W4226474429 hasConceptScore W4226474429C173801870 @default.
- W4226474429 hasConceptScore W4226474429C2776214188 @default.
- W4226474429 hasConceptScore W4226474429C2779127903 @default.
- W4226474429 hasConceptScore W4226474429C41008148 @default.
- W4226474429 hasConceptScore W4226474429C49937458 @default.
- W4226474429 hasConceptScore W4226474429C50644808 @default.
- W4226474429 hasFunder F4320322120 @default.
- W4226474429 hasFunder F4320338335 @default.
- W4226474429 hasIssue "5" @default.
- W4226474429 hasLocation W42264744291 @default.
- W4226474429 hasLocation W42264744292 @default.
- W4226474429 hasLocation W42264744293 @default.
- W4226474429 hasOpenAccess W4226474429 @default.
- W4226474429 hasPrimaryLocation W42264744291 @default.
- W4226474429 hasRelatedWork W2961085424 @default.
- W4226474429 hasRelatedWork W3046775127 @default.
- W4226474429 hasRelatedWork W3170094116 @default.
- W4226474429 hasRelatedWork W4205958290 @default.
- W4226474429 hasRelatedWork W4285260836 @default.
- W4226474429 hasRelatedWork W4286629047 @default.
- W4226474429 hasRelatedWork W4290792893 @default.
- W4226474429 hasRelatedWork W4306321456 @default.
- W4226474429 hasRelatedWork W4306674287 @default.
- W4226474429 hasRelatedWork W4224009465 @default.
- W4226474429 hasVolume "33" @default.
- W4226474429 isParatext "false" @default.
- W4226474429 isRetracted "false" @default.
- W4226474429 workType "article" @default.