FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2965330733> ?p ?o ?g. }

• W2965330733 abstract "Variational Bayes (VB) has become a widely-used tool for Bayesian inference in statistics and machine learning. Nonetheless, the development of the existing VB algorithms is so far generally restricted to the case where the variational parameter space is Euclidean, which hinders the potential broad application of VB methods. This paper extends the scope of VB to the case where the variational parameter space is a Riemannian manifold. We develop an efficient manifold-based VB algorithm that exploits both the geometric structure of the constraint parameter space and the information geometry of the manifold of VB approximating probability distributions. Our algorithm is provably convergent and achieves a convergence rate of order $mathcal O(1/sqrt{T})$ and $mathcal O(1/T^{2-2epsilon})$ for a non-convex evidence lower bound function and a strongly retraction-convex evidence lower bound function, respectively. We develop in particular two manifold VB algorithms, Manifold Gaussian VB and Manifold Neural Net VB, and demonstrate through numerical experiments that the proposed algorithms are stable, less sensitive to initialization and compares favourably to existing VB methods." @default.
• W2965330733 created "2019-08-13" @default.
• W2965330733 creator A5014870015 @default.
• W2965330733 creator A5040875924 @default.
• W2965330733 creator A5051847219 @default.
• W2965330733 date "2019-08-08" @default.
• W2965330733 modified "2023-10-18" @default.
• W2965330733 title "Variational Bayes on Manifolds" @default.
• W2965330733 cites W1479979375 @default.
• W2965330733 cites W1508953807 @default.
• W2965330733 cites W1513873506 @default.
• W2965330733 cites W1516111018 @default.
• W2965330733 cites W1528905581 @default.
• W2965330733 cites W1628278711 @default.
• W2965330733 cites W1804110266 @default.
• W2965330733 cites W1974766144 @default.
• W2965330733 cites W1999996900 @default.
• W2965330733 cites W2011981966 @default.
• W2965330733 cites W2045512849 @default.
• W2965330733 cites W2059424427 @default.
• W2965330733 cites W2060624994 @default.
• W2965330733 cites W2089465567 @default.
• W2965330733 cites W2105658140 @default.
• W2965330733 cites W2108907664 @default.
• W2965330733 cites W2140660104 @default.
• W2965330733 cites W2155894447 @default.
• W2965330733 cites W2156529323 @default.
• W2965330733 cites W2166851633 @default.
• W2965330733 cites W2167623372 @default.
• W2965330733 cites W2171911691 @default.
• W2965330733 cites W2224136389 @default.
• W2965330733 cites W2555456875 @default.
• W2965330733 cites W2579661004 @default.
• W2965330733 cites W260529710 @default.
• W2965330733 cites W2777072123 @default.
• W2965330733 cites W2791004381 @default.
• W2965330733 cites W2901030372 @default.
• W2965330733 cites W2936560958 @default.
• W2965330733 cites W2948034225 @default.
• W2965330733 cites W2950085493 @default.
• W2965330733 cites W2951004968 @default.
• W2965330733 cites W2962791800 @default.
• W2965330733 cites W2963991960 @default.
• W2965330733 cites W2964050265 @default.
• W2965330733 cites W3086499488 @default.
• W2965330733 cites W3102317997 @default.
• W2965330733 cites W3157881463 @default.
• W2965330733 cites W95577512 @default.
• W2965330733 hasPublicationYear "2019" @default.
• W2965330733 type Work @default.
• W2965330733 sameAs 2965330733 @default.
• W2965330733 citedByCount "7" @default.
• W2965330733 countsByYear W29653307332019 @default.
• W2965330733 countsByYear W29653307332020 @default.
• W2965330733 countsByYear W29653307332021 @default.
• W2965330733 crossrefType "posted-content" @default.
• W2965330733 hasAuthorship W2965330733A5014870015 @default.
• W2965330733 hasAuthorship W2965330733A5040875924 @default.
• W2965330733 hasAuthorship W2965330733A5051847219 @default.
• W2965330733 hasConcept C107673813 @default.
• W2965330733 hasConcept C111919701 @default.
• W2965330733 hasConcept C11413529 @default.
• W2965330733 hasConcept C114466953 @default.
• W2965330733 hasConcept C114614502 @default.
• W2965330733 hasConcept C127413603 @default.
• W2965330733 hasConcept C14036430 @default.
• W2965330733 hasConcept C142730499 @default.
• W2965330733 hasConcept C154945302 @default.
• W2965330733 hasConcept C184720557 @default.
• W2965330733 hasConcept C186450821 @default.
• W2965330733 hasConcept C199360897 @default.
• W2965330733 hasConcept C202444582 @default.
• W2965330733 hasConcept C207201462 @default.
• W2965330733 hasConcept C2524010 @default.
• W2965330733 hasConcept C2778572836 @default.
• W2965330733 hasConcept C28826006 @default.
• W2965330733 hasConcept C33923547 @default.
• W2965330733 hasConcept C41008148 @default.
• W2965330733 hasConcept C529865628 @default.
• W2965330733 hasConcept C73586568 @default.
• W2965330733 hasConcept C78458016 @default.
• W2965330733 hasConcept C78519656 @default.
• W2965330733 hasConcept C86803240 @default.
• W2965330733 hasConceptScore W2965330733C107673813 @default.
• W2965330733 hasConceptScore W2965330733C111919701 @default.
• W2965330733 hasConceptScore W2965330733C11413529 @default.
• W2965330733 hasConceptScore W2965330733C114466953 @default.
• W2965330733 hasConceptScore W2965330733C114614502 @default.
• W2965330733 hasConceptScore W2965330733C127413603 @default.
• W2965330733 hasConceptScore W2965330733C14036430 @default.
• W2965330733 hasConceptScore W2965330733C142730499 @default.
• W2965330733 hasConceptScore W2965330733C154945302 @default.
• W2965330733 hasConceptScore W2965330733C184720557 @default.
• W2965330733 hasConceptScore W2965330733C186450821 @default.
• W2965330733 hasConceptScore W2965330733C199360897 @default.
• W2965330733 hasConceptScore W2965330733C202444582 @default.
• W2965330733 hasConceptScore W2965330733C207201462 @default.
• W2965330733 hasConceptScore W2965330733C2524010 @default.
• W2965330733 hasConceptScore W2965330733C2778572836 @default.
• W2965330733 hasConceptScore W2965330733C28826006 @default.

• next