FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2922007432> ?p ?o ?g. }

• W2922007432 abstract "Author(s): Lou, Qi | Advisor(s): Ihler, Alexander | Abstract: Graphical models are a powerful framework for modeling interactions within complex systems. Reasoning over graphical models typically involves answering inference queries, such as computing the most likely configuration (maximum a posteriori or MAP) or evaluating the marginals or normalizing constant of a distribution (the partition function); a task called marginal MAP generalizes these two by maximizing over a subset of variables while marginalizing over the rest.Exact computation of these queries is known to be intractable in general, leading to the development of many approximate schemes, the major categories of which are variational methods, search algorithms, and Monte Carlo sampling. Within these, anytime techniques that provide some guarantees on the correct value, and can be improved with more computational effort, are valued for quickly providing users with confidence intervals or certificates of accuracy and allow users to decide the desired balance of quality, time and memory.In this dissertation, we develop a series of approximate inference algorithms for the partition function and marginal MAP with anytime properties by leveraging ideas and techniques from the three inference paradigms, and integrating them to provide hybrid solutions that inherit the strengths of all three approaches. We propose anytime anyspace best-first search algorithms that provide deterministic bounds on the partition function and marginal MAP. These best-first search schemes take advantage of both AND/OR tree search and optimized variational heuristics. We then extend this approach to give anytime probabilistic confidence bounds via a dynamic importance sampling algorithm, which interleaves importance sampling (using proposal distributions extracted from the variational bound) with our best-first search algorithm to refine the proposal. We also propose a framework for interleaving sampling with the optimization of the initial variational bound, which can automatically balance its computational effort between the two schemes. Overall, we show that our hybrid algorithms perform significantly better than existing methods, giving flexible approaches with excellent anytime confidence bounds." @default.
• W2922007432 created "2019-03-22" @default.
• W2922007432 creator A5077858309 @default.
• W2922007432 date "2018-01-01" @default.
• W2922007432 modified "2023-09-26" @default.
• W2922007432 title "Anytime Approximate Inference in Graphical Models" @default.
• W2922007432 cites W1480349804 @default.
• W2922007432 cites W1489119587 @default.
• W2922007432 cites W1491239593 @default.
• W2922007432 cites W1505356468 @default.
• W2922007432 cites W1505442699 @default.
• W2922007432 cites W1510302714 @default.
• W2922007432 cites W1511187233 @default.
• W2922007432 cites W1511986666 @default.
• W2922007432 cites W1518630026 @default.
• W2922007432 cites W1520781524 @default.
• W2922007432 cites W1535430927 @default.
• W2922007432 cites W1546981760 @default.
• W2922007432 cites W1550892302 @default.
• W2922007432 cites W1601035521 @default.
• W2922007432 cites W1697053819 @default.
• W2922007432 cites W1748004346 @default.
• W2922007432 cites W177438173 @default.
• W2922007432 cites W1934021597 @default.
• W2922007432 cites W1969483458 @default.
• W2922007432 cites W1982333717 @default.
• W2922007432 cites W1999901857 @default.
• W2922007432 cites W2006912660 @default.
• W2922007432 cites W2008740515 @default.
• W2922007432 cites W2022637125 @default.
• W2922007432 cites W2062818807 @default.
• W2922007432 cites W2071988465 @default.
• W2922007432 cites W2095999997 @default.
• W2922007432 cites W2096192494 @default.
• W2922007432 cites W2100034662 @default.
• W2922007432 cites W2101968469 @default.
• W2922007432 cites W2103012681 @default.
• W2922007432 cites W2106317217 @default.
• W2922007432 cites W2112796928 @default.
• W2922007432 cites W2114267503 @default.
• W2922007432 cites W2120340025 @default.
• W2922007432 cites W2121863487 @default.
• W2922007432 cites W2122003883 @default.
• W2922007432 cites W2125186487 @default.
• W2922007432 cites W2125631472 @default.
• W2922007432 cites W2126316555 @default.
• W2922007432 cites W2130178369 @default.
• W2922007432 cites W2130467648 @default.
• W2922007432 cites W2135094946 @default.
• W2922007432 cites W2135194391 @default.
• W2922007432 cites W2136064009 @default.
• W2922007432 cites W2137822088 @default.
• W2922007432 cites W2150470619 @default.
• W2922007432 cites W2153806142 @default.
• W2922007432 cites W2154055561 @default.
• W2922007432 cites W2163316275 @default.
• W2922007432 cites W2164568295 @default.
• W2922007432 cites W2171035569 @default.
• W2922007432 cites W2174332830 @default.
• W2922007432 cites W2182427043 @default.
• W2922007432 cites W2183877918 @default.
• W2922007432 cites W223130948 @default.
• W2922007432 cites W2253548355 @default.
• W2922007432 cites W2407138673 @default.
• W2922007432 cites W2408128689 @default.
• W2922007432 cites W2410190751 @default.
• W2922007432 cites W2566054401 @default.
• W2922007432 cites W2572061353 @default.
• W2922007432 cites W2605178440 @default.
• W2922007432 cites W2622268367 @default.
• W2922007432 cites W2751676477 @default.
• W2922007432 cites W2787531831 @default.
• W2922007432 cites W2808602246 @default.
• W2922007432 cites W2809278324 @default.
• W2922007432 cites W2904033531 @default.
• W2922007432 cites W2912526555 @default.
• W2922007432 cites W2913554033 @default.
• W2922007432 cites W2949395086 @default.
• W2922007432 cites W2952143373 @default.
• W2922007432 cites W2952569054 @default.
• W2922007432 cites W2952886082 @default.
• W2922007432 cites W2963848397 @default.
• W2922007432 cites W2964077009 @default.
• W2922007432 cites W3021035068 @default.
• W2922007432 cites W3023084835 @default.
• W2922007432 hasPublicationYear "2018" @default.
• W2922007432 type Work @default.
• W2922007432 sameAs 2922007432 @default.
• W2922007432 citedByCount "0" @default.
• W2922007432 crossrefType "journal-article" @default.
• W2922007432 hasAuthorship W2922007432A5077858309 @default.
• W2922007432 hasConcept C105795698 @default.
• W2922007432 hasConcept C111919701 @default.
• W2922007432 hasConcept C11413529 @default.
• W2922007432 hasConcept C127705205 @default.
• W2922007432 hasConcept C154945302 @default.
• W2922007432 hasConcept C155846161 @default.
• W2922007432 hasConcept C169272836 @default.
• W2922007432 hasConcept C2776214188 @default.
• W2922007432 hasConcept C2777472644 @default.

• next