Matches in SemOpenAlex for { <https://semopenalex.org/work/W1990147185> ?p ?o ?g. }
- W1990147185 endingPage "1451" @default.
- W1990147185 startingPage "1393" @default.
- W1990147185 abstract "Neural associative memories are perceptron-like single-layer networks with fast synaptic learning typically storing discrete associations between pairs of neural activity patterns. Previous work optimized the memory capacity for various models of synaptic learning: linear Hopfield-type rules, the Willshaw model employing binary synapses, or the BCPNN rule of Lansner and Ekeberg, for example. Here I show that all of these previous models are limit cases of a general optimal model where synaptic learning is determined by probabilistic Bayesian considerations. Asymptotically, for large networks and very sparse neuron activity, the Bayesian model becomes identical to an inhibitory implementation of the Willshaw and BCPNN-type models. For less sparse patterns, the Bayesian model becomes identical to Hopfield-type networks employing the covariance rule. For intermediate sparseness or finite networks, the optimal Bayesian learning rule differs from the previous models and can significantly improve memory performance. I also provide a unified analytical framework to determine memory capacity at a given output noise level that links approaches based on mutual information, Hamming distance, and signal-to-noise ratio." @default.
- W1990147185 created "2016-06-24" @default.
- W1990147185 creator A5032109856 @default.
- W1990147185 date "2011-06-01" @default.
- W1990147185 modified "2023-10-01" @default.
- W1990147185 title "Neural Associative Memory with Optimal Bayesian Learning" @default.
- W1990147185 cites W139929112 @default.
- W1990147185 cites W1514853588 @default.
- W1990147185 cites W1583817925 @default.
- W1990147185 cites W1605844890 @default.
- W1990147185 cites W1666326031 @default.
- W1990147185 cites W1729998890 @default.
- W1990147185 cites W1964049964 @default.
- W1990147185 cites W1973144389 @default.
- W1990147185 cites W1982131418 @default.
- W1990147185 cites W1986857526 @default.
- W1990147185 cites W1994811745 @default.
- W1990147185 cites W1995142449 @default.
- W1990147185 cites W1996253558 @default.
- W1990147185 cites W1999571566 @default.
- W1990147185 cites W2001570872 @default.
- W1990147185 cites W2003952197 @default.
- W1990147185 cites W2006327398 @default.
- W1990147185 cites W2015418956 @default.
- W1990147185 cites W2019165547 @default.
- W1990147185 cites W2022442596 @default.
- W1990147185 cites W2024418895 @default.
- W1990147185 cites W2025603247 @default.
- W1990147185 cites W2027527273 @default.
- W1990147185 cites W2028281435 @default.
- W1990147185 cites W2028744261 @default.
- W1990147185 cites W2030450972 @default.
- W1990147185 cites W2037195447 @default.
- W1990147185 cites W2038307386 @default.
- W1990147185 cites W2042469993 @default.
- W1990147185 cites W2049165798 @default.
- W1990147185 cites W2051992922 @default.
- W1990147185 cites W2053078562 @default.
- W1990147185 cites W2057529293 @default.
- W1990147185 cites W2058960948 @default.
- W1990147185 cites W2064050145 @default.
- W1990147185 cites W2075187489 @default.
- W1990147185 cites W2076551955 @default.
- W1990147185 cites W2077116346 @default.
- W1990147185 cites W2077501133 @default.
- W1990147185 cites W2077680606 @default.
- W1990147185 cites W2078516287 @default.
- W1990147185 cites W2081118181 @default.
- W1990147185 cites W2082622165 @default.
- W1990147185 cites W2085415665 @default.
- W1990147185 cites W2088514698 @default.
- W1990147185 cites W2099111195 @default.
- W1990147185 cites W2103019454 @default.
- W1990147185 cites W2103744641 @default.
- W1990147185 cites W2104551188 @default.
- W1990147185 cites W2110016065 @default.
- W1990147185 cites W2119051448 @default.
- W1990147185 cites W2125001634 @default.
- W1990147185 cites W2126076951 @default.
- W1990147185 cites W2128084896 @default.
- W1990147185 cites W2129217160 @default.
- W1990147185 cites W2129386380 @default.
- W1990147185 cites W2133146712 @default.
- W1990147185 cites W2133943581 @default.
- W1990147185 cites W2135576811 @default.
- W1990147185 cites W2140785063 @default.
- W1990147185 cites W2154757610 @default.
- W1990147185 cites W2155051950 @default.
- W1990147185 cites W2161144680 @default.
- W1990147185 cites W2161389622 @default.
- W1990147185 cites W2161496668 @default.
- W1990147185 cites W2213243153 @default.
- W1990147185 cites W2321192406 @default.
- W1990147185 cites W3178142071 @default.
- W1990147185 cites W4211174381 @default.
- W1990147185 cites W4212941528 @default.
- W1990147185 cites W4231543040 @default.
- W1990147185 cites W4232599013 @default.
- W1990147185 cites W4237454176 @default.
- W1990147185 cites W4240778903 @default.
- W1990147185 cites W4241400578 @default.
- W1990147185 cites W4376595930 @default.
- W1990147185 cites W1496255390 @default.
- W1990147185 cites W2108962376 @default.
- W1990147185 doi "https://doi.org/10.1162/neco_a_00127" @default.
- W1990147185 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/21395440" @default.
- W1990147185 hasPublicationYear "2011" @default.
- W1990147185 type Work @default.
- W1990147185 sameAs 1990147185 @default.
- W1990147185 citedByCount "27" @default.
- W1990147185 countsByYear W19901471852012 @default.
- W1990147185 countsByYear W19901471852013 @default.
- W1990147185 countsByYear W19901471852014 @default.
- W1990147185 countsByYear W19901471852015 @default.
- W1990147185 countsByYear W19901471852016 @default.
- W1990147185 countsByYear W19901471852017 @default.
- W1990147185 countsByYear W19901471852018 @default.
- W1990147185 countsByYear W19901471852019 @default.