FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2945431146> ?p ?o ?g. }

• W2945431146 abstract "Sum-product networks (SPNs) are flexible density estimators and have received significant attention due to their attractive inference properties. While parameter learning in SPNs is well developed, structure learning leaves something to be desired: Even though there is a plethora of SPN structure learners, most of them are somewhat ad-hoc and based on intuition rather than a clear learning principle. In this paper, we introduce a well-principled Bayesian framework for SPN structure learning. First, we decompose the problem into i) laying out a computational graph, and ii) learning the so-called scope function over the graph. The first is rather unproblematic and akin to neural network architecture validation. The second represents the effective structure of the SPN and needs to respect the usual structural constraints in SPN, i.e. completeness and decomposability. While representing and learning the scope function is somewhat involved in general, in this paper, we propose a natural parametrisation for an important and widely used special case of SPNs. These structural parameters are incorporated into a Bayesian model, such that simultaneous structure and parameter learning is cast into monolithic Bayesian posterior inference. In various experiments, our Bayesian SPNs often improve test likelihoods over greedy SPN learners. Further, since the Bayesian framework protects against overfitting, we can evaluate hyper-parameters directly on the Bayesian model score, waiving the need for a separate validation set, which is especially beneficial in low data regimes. Bayesian SPNs can be applied to heterogeneous domains and can easily be extended to nonparametric formulations. Moreover, our Bayesian approach is the first, which consistently and robustly learns SPN structures under missing data." @default.
• W2945431146 created "2019-05-29" @default.
• W2945431146 creator A5002907320 @default.
• W2945431146 creator A5010820865 @default.
• W2945431146 creator A5015798259 @default.
• W2945431146 creator A5057485787 @default.
• W2945431146 creator A5060407332 @default.
• W2945431146 date "2019-05-26" @default.
• W2945431146 modified "2023-10-01" @default.
• W2945431146 title "Bayesian Learning of Sum-Product Networks" @default.
• W2945431146 cites W142410360 @default.
• W2945431146 cites W1511986666 @default.
• W2945431146 cites W1521578411 @default.
• W2945431146 cites W1767574678 @default.
• W2945431146 cites W1835282959 @default.
• W2945431146 cites W1865606400 @default.
• W2945431146 cites W1972978214 @default.
• W2945431146 cites W2008906462 @default.
• W2945431146 cites W2119900738 @default.
• W2945431146 cites W2128744540 @default.
• W2945431146 cites W2140251433 @default.
• W2945431146 cites W2144731007 @default.
• W2945431146 cites W2159687189 @default.
• W2945431146 cites W2168906135 @default.
• W2945431146 cites W2170112109 @default.
• W2945431146 cites W2182706877 @default.
• W2945431146 cites W2196348858 @default.
• W2945431146 cites W223130948 @default.
• W2945431146 cites W2280929120 @default.
• W2945431146 cites W2394620536 @default.
• W2945431146 cites W2395740735 @default.
• W2945431146 cites W2399803916 @default.
• W2945431146 cites W2416891345 @default.
• W2945431146 cites W2420777250 @default.
• W2945431146 cites W2496114304 @default.
• W2945431146 cites W2534135591 @default.
• W2945431146 cites W2592688734 @default.
• W2945431146 cites W2734342344 @default.
• W2945431146 cites W2788951710 @default.
• W2945431146 cites W2807461489 @default.
• W2945431146 cites W283954808 @default.
• W2945431146 cites W2890008096 @default.
• W2945431146 cites W2899000040 @default.
• W2945431146 cites W2904464104 @default.
• W2945431146 cites W2910209909 @default.
• W2945431146 cites W2945132577 @default.
• W2945431146 cites W2946492213 @default.
• W2945431146 cites W2946666653 @default.
• W2945431146 cites W2956476286 @default.
• W2945431146 cites W2963276753 @default.
• W2945431146 cites W2963368172 @default.
• W2945431146 cites W2963374479 @default.
• W2945431146 cites W2963551771 @default.
• W2945431146 cites W2963818173 @default.
• W2945431146 cites W2964274470 @default.
• W2945431146 cites W2965519836 @default.
• W2945431146 cites W2781363082 @default.
• W2945431146 cites W2898877291 @default.
• W2945431146 cites W3145738572 @default.
• W2945431146 hasPublicationYear "2019" @default.
• W2945431146 type Work @default.
• W2945431146 sameAs 2945431146 @default.
• W2945431146 citedByCount "2" @default.
• W2945431146 countsByYear W29454311462019 @default.
• W2945431146 countsByYear W29454311462020 @default.
• W2945431146 crossrefType "posted-content" @default.
• W2945431146 hasAuthorship W2945431146A5002907320 @default.
• W2945431146 hasAuthorship W2945431146A5010820865 @default.
• W2945431146 hasAuthorship W2945431146A5015798259 @default.
• W2945431146 hasAuthorship W2945431146A5057485787 @default.
• W2945431146 hasAuthorship W2945431146A5060407332 @default.
• W2945431146 hasConcept C101112237 @default.
• W2945431146 hasConcept C107673813 @default.
• W2945431146 hasConcept C119857082 @default.
• W2945431146 hasConcept C132525143 @default.
• W2945431146 hasConcept C154945302 @default.
• W2945431146 hasConcept C160234255 @default.
• W2945431146 hasConcept C22019652 @default.
• W2945431146 hasConcept C2776214188 @default.
• W2945431146 hasConcept C33724603 @default.
• W2945431146 hasConcept C41008148 @default.
• W2945431146 hasConcept C50644808 @default.
• W2945431146 hasConcept C71983512 @default.
• W2945431146 hasConcept C80444323 @default.
• W2945431146 hasConceptScore W2945431146C101112237 @default.
• W2945431146 hasConceptScore W2945431146C107673813 @default.
• W2945431146 hasConceptScore W2945431146C119857082 @default.
• W2945431146 hasConceptScore W2945431146C132525143 @default.
• W2945431146 hasConceptScore W2945431146C154945302 @default.
• W2945431146 hasConceptScore W2945431146C160234255 @default.
• W2945431146 hasConceptScore W2945431146C22019652 @default.
• W2945431146 hasConceptScore W2945431146C2776214188 @default.
• W2945431146 hasConceptScore W2945431146C33724603 @default.
• W2945431146 hasConceptScore W2945431146C41008148 @default.
• W2945431146 hasConceptScore W2945431146C50644808 @default.
• W2945431146 hasConceptScore W2945431146C71983512 @default.
• W2945431146 hasConceptScore W2945431146C80444323 @default.
• W2945431146 hasLocation W29454311461 @default.
• W2945431146 hasOpenAccess W2945431146 @default.
• W2945431146 hasPrimaryLocation W29454311461 @default.

• next