Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386002429> ?p ?o ?g. }
- W4386002429 endingPage "129143" @default.
- W4386002429 startingPage "129143" @default.
- W4386002429 abstract "We consider dense, associative neural-networks trained with no supervision and we investigate their computational capabilities analytically, via statistical-mechanics tools, and numerically, via Monte Carlo simulations. In particular, we obtain a phase diagram summarizing their performance as a function of the control parameters (e.g. quality and quantity of the training dataset, network storage, noise) that is valid in the limit of large network size and structureless datasets. Moreover, we establish a bridge between macroscopic observables standardly used in statistical mechanics and loss functions typically used in the machine learning. As technical remarks, from the analytical side, we extend Guerra’s interpolation to tackle the non-Gaussian distributions involved in the post-synaptic potentials while, from the computational counterpart, we insert Plefka’s approximation in the Monte Carlo scheme, to speed up the evaluation of the synaptic tensor, overall obtaining a novel and broad approach to investigate unsupervised learning in neural networks, beyond the shallow limit." @default.
- W4386002429 created "2023-08-20" @default.
- W4386002429 creator A5005761274 @default.
- W4386002429 creator A5011672584 @default.
- W4386002429 creator A5024505700 @default.
- W4386002429 creator A5028701946 @default.
- W4386002429 creator A5041457930 @default.
- W4386002429 creator A5044898565 @default.
- W4386002429 creator A5053268916 @default.
- W4386002429 creator A5077516005 @default.
- W4386002429 date "2023-10-01" @default.
- W4386002429 modified "2023-10-07" @default.
- W4386002429 title "Dense Hebbian neural networks: A replica symmetric picture of unsupervised learning" @default.
- W4386002429 cites W105622892 @default.
- W4386002429 cites W1661843026 @default.
- W4386002429 cites W1985452037 @default.
- W4386002429 cites W2027341885 @default.
- W4386002429 cites W2029667457 @default.
- W4386002429 cites W2036487714 @default.
- W4386002429 cites W2059988607 @default.
- W4386002429 cites W2069097920 @default.
- W4386002429 cites W2075503349 @default.
- W4386002429 cites W2080792322 @default.
- W4386002429 cites W2087636648 @default.
- W4386002429 cites W2128084896 @default.
- W4386002429 cites W2161410247 @default.
- W4386002429 cites W2226502241 @default.
- W4386002429 cites W2923537029 @default.
- W4386002429 cites W2963305853 @default.
- W4386002429 cites W2963809228 @default.
- W4386002429 cites W2964102258 @default.
- W4386002429 cites W2979858600 @default.
- W4386002429 cites W2989716259 @default.
- W4386002429 cites W3015748378 @default.
- W4386002429 cites W3027269453 @default.
- W4386002429 cites W3029326706 @default.
- W4386002429 cites W3049152644 @default.
- W4386002429 cites W3097007160 @default.
- W4386002429 cites W3099406601 @default.
- W4386002429 cites W3100092254 @default.
- W4386002429 cites W3104469760 @default.
- W4386002429 cites W3105463048 @default.
- W4386002429 cites W3112377917 @default.
- W4386002429 cites W3185176267 @default.
- W4386002429 cites W3212757835 @default.
- W4386002429 cites W4210620801 @default.
- W4386002429 cites W4225307025 @default.
- W4386002429 cites W4296104280 @default.
- W4386002429 cites W4307644102 @default.
- W4386002429 cites W4309866921 @default.
- W4386002429 doi "https://doi.org/10.1016/j.physa.2023.129143" @default.
- W4386002429 hasPublicationYear "2023" @default.
- W4386002429 type Work @default.
- W4386002429 citedByCount "0" @default.
- W4386002429 crossrefType "journal-article" @default.
- W4386002429 hasAuthorship W4386002429A5005761274 @default.
- W4386002429 hasAuthorship W4386002429A5011672584 @default.
- W4386002429 hasAuthorship W4386002429A5024505700 @default.
- W4386002429 hasAuthorship W4386002429A5028701946 @default.
- W4386002429 hasAuthorship W4386002429A5041457930 @default.
- W4386002429 hasAuthorship W4386002429A5044898565 @default.
- W4386002429 hasAuthorship W4386002429A5053268916 @default.
- W4386002429 hasAuthorship W4386002429A5077516005 @default.
- W4386002429 hasBestOaLocation W43860024292 @default.
- W4386002429 hasConcept C104114177 @default.
- W4386002429 hasConcept C105795698 @default.
- W4386002429 hasConcept C111437709 @default.
- W4386002429 hasConcept C11413529 @default.
- W4386002429 hasConcept C119857082 @default.
- W4386002429 hasConcept C121332964 @default.
- W4386002429 hasConcept C121864883 @default.
- W4386002429 hasConcept C134306372 @default.
- W4386002429 hasConcept C137800194 @default.
- W4386002429 hasConcept C142362112 @default.
- W4386002429 hasConcept C151201525 @default.
- W4386002429 hasConcept C153349607 @default.
- W4386002429 hasConcept C154945302 @default.
- W4386002429 hasConcept C192576344 @default.
- W4386002429 hasConcept C19499675 @default.
- W4386002429 hasConcept C2775937380 @default.
- W4386002429 hasConcept C33923547 @default.
- W4386002429 hasConcept C41008148 @default.
- W4386002429 hasConcept C50644808 @default.
- W4386002429 hasConcept C8038995 @default.
- W4386002429 hasConcept C99874945 @default.
- W4386002429 hasConceptScore W4386002429C104114177 @default.
- W4386002429 hasConceptScore W4386002429C105795698 @default.
- W4386002429 hasConceptScore W4386002429C111437709 @default.
- W4386002429 hasConceptScore W4386002429C11413529 @default.
- W4386002429 hasConceptScore W4386002429C119857082 @default.
- W4386002429 hasConceptScore W4386002429C121332964 @default.
- W4386002429 hasConceptScore W4386002429C121864883 @default.
- W4386002429 hasConceptScore W4386002429C134306372 @default.
- W4386002429 hasConceptScore W4386002429C137800194 @default.
- W4386002429 hasConceptScore W4386002429C142362112 @default.
- W4386002429 hasConceptScore W4386002429C151201525 @default.
- W4386002429 hasConceptScore W4386002429C153349607 @default.
- W4386002429 hasConceptScore W4386002429C154945302 @default.