FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2105780862> ?p ?o ?g. }

• W2105780862 endingPage "1412" @default.
• W2105780862 startingPage "1380" @default.
• W2105780862 abstract "Currently all important, low-level, unsupervised network learning algorithms follow the paradigm of Hebb, where input and output activity are correlated to change the connection strength of a synapse. However, as a consequence, classical Hebbian learning always carries a potentially destabilizing autocorrelation term, which is due to the fact that every input is in a weighted form reflected in the neuron's output. This self-correlation can lead to positive feedback, where increasing weights will increase the output, and vice versa, which may result in divergence. This can be avoided by different strategies like weight normalization or weight saturation, which, however, can cause different problems. Consequently, in most cases, high learning rates cannot be used for Hebbian learning, leading to relatively slow convergence. Here we introduce a novel correlation-based learning rule that is related to our isotropic sequence order (ISO) learning rule (Porr & Wörgötter, 2003a), but replaces the derivative of the output in the learning rule with the derivative of the reflex input. Hence, the new rule uses input correlations only, effectively implementing strict heterosynaptic learning. This looks like a minor modification but leads to dramatically improved properties. Elimination of the output from the learning rule removes the unwanted, destabilizing autocorrelation term, allowing us to use high learning rates. As a consequence, we can mathematically show that the theoretical optimum of one-shot learning can be reached under ideal conditions with the new rule. This result is then tested against four different experimental setups, and we will show that in all of them, very few (and sometimes only one) learning experiences are needed to achieve the learning goal. As a consequence, the new learning rule is up to 100 times faster and in general more stable than ISO learning." @default.
• W2105780862 created "2016-06-24" @default.
• W2105780862 creator A5023811677 @default.
• W2105780862 creator A5091850362 @default.
• W2105780862 date "2006-06-01" @default.
• W2105780862 modified "2023-10-16" @default.
• W2105780862 title "Strongly Improved Stability and Faster Convergence of Temporal Sequence Learning by Using Input Correlations Only" @default.
• W2105780862 cites W1501143146 @default.
• W2105780862 cites W1542305729 @default.
• W2105780862 cites W1733248925 @default.
• W2105780862 cites W1978579256 @default.
• W2105780862 cites W1981698940 @default.
• W2105780862 cites W1983809482 @default.
• W2105780862 cites W2006327398 @default.
• W2105780862 cites W2009782005 @default.
• W2105780862 cites W2017910024 @default.
• W2105780862 cites W2024488143 @default.
• W2105780862 cites W2031688164 @default.
• W2105780862 cites W2040598998 @default.
• W2105780862 cites W2040870580 @default.
• W2105780862 cites W2042322703 @default.
• W2105780862 cites W2048250261 @default.
• W2105780862 cites W2065546903 @default.
• W2105780862 cites W2067623998 @default.
• W2105780862 cites W2084339020 @default.
• W2105780862 cites W2084702088 @default.
• W2105780862 cites W2097872145 @default.
• W2105780862 cites W2099652807 @default.
• W2105780862 cites W2107726111 @default.
• W2105780862 cites W2110724392 @default.
• W2105780862 cites W2117726420 @default.
• W2105780862 cites W2122925692 @default.
• W2105780862 cites W2125200267 @default.
• W2105780862 cites W2136621137 @default.
• W2105780862 cites W2137290292 @default.
• W2105780862 cites W2150863395 @default.
• W2105780862 cites W2164936082 @default.
• W2105780862 cites W2169959530 @default.
• W2105780862 cites W2397253692 @default.
• W2105780862 cites W2432567885 @default.
• W2105780862 cites W2887242076 @default.
• W2105780862 cites W3041202696 @default.
• W2105780862 cites W4248400504 @default.
• W2105780862 cites W4299511277 @default.
• W2105780862 cites W2185212694 @default.
• W2105780862 doi "https://doi.org/10.1162/neco.2006.18.6.1380" @default.
• W2105780862 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/16764508" @default.
• W2105780862 hasPublicationYear "2006" @default.
• W2105780862 type Work @default.
• W2105780862 sameAs 2105780862 @default.
• W2105780862 citedByCount "63" @default.
• W2105780862 countsByYear W21057808622012 @default.
• W2105780862 countsByYear W21057808622013 @default.
• W2105780862 countsByYear W21057808622014 @default.
• W2105780862 countsByYear W21057808622016 @default.
• W2105780862 countsByYear W21057808622017 @default.
• W2105780862 countsByYear W21057808622018 @default.
• W2105780862 countsByYear W21057808622019 @default.
• W2105780862 countsByYear W21057808622020 @default.
• W2105780862 countsByYear W21057808622021 @default.
• W2105780862 countsByYear W21057808622022 @default.
• W2105780862 countsByYear W21057808622023 @default.
• W2105780862 crossrefType "journal-article" @default.
• W2105780862 hasAuthorship W2105780862A5023811677 @default.
• W2105780862 hasAuthorship W2105780862A5091850362 @default.
• W2105780862 hasConcept C105795698 @default.
• W2105780862 hasConcept C111437709 @default.
• W2105780862 hasConcept C112972136 @default.
• W2105780862 hasConcept C11413529 @default.
• W2105780862 hasConcept C119857082 @default.
• W2105780862 hasConcept C154945302 @default.
• W2105780862 hasConcept C2778112365 @default.
• W2105780862 hasConcept C2779127903 @default.
• W2105780862 hasConcept C33923547 @default.
• W2105780862 hasConcept C40506919 @default.
• W2105780862 hasConcept C41008148 @default.
• W2105780862 hasConcept C50644808 @default.
• W2105780862 hasConcept C5297727 @default.
• W2105780862 hasConcept C54355233 @default.
• W2105780862 hasConcept C86803240 @default.
• W2105780862 hasConceptScore W2105780862C105795698 @default.
• W2105780862 hasConceptScore W2105780862C111437709 @default.
• W2105780862 hasConceptScore W2105780862C112972136 @default.
• W2105780862 hasConceptScore W2105780862C11413529 @default.
• W2105780862 hasConceptScore W2105780862C119857082 @default.
• W2105780862 hasConceptScore W2105780862C154945302 @default.
• W2105780862 hasConceptScore W2105780862C2778112365 @default.
• W2105780862 hasConceptScore W2105780862C2779127903 @default.
• W2105780862 hasConceptScore W2105780862C33923547 @default.
• W2105780862 hasConceptScore W2105780862C40506919 @default.
• W2105780862 hasConceptScore W2105780862C41008148 @default.
• W2105780862 hasConceptScore W2105780862C50644808 @default.
• W2105780862 hasConceptScore W2105780862C5297727 @default.
• W2105780862 hasConceptScore W2105780862C54355233 @default.
• W2105780862 hasConceptScore W2105780862C86803240 @default.
• W2105780862 hasIssue "6" @default.
• W2105780862 hasLocation W21057808621 @default.
• W2105780862 hasLocation W21057808622 @default.

• next