FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1830759839> ?p ?o ?g. }

• W1830759839 abstract "Author(s): Critch, Andrew | Advisor(s): Sturmfels, Bernd | Abstract: This thesis embodies a collection of algebraic techniques and results on hidden Markov models, related models in quantum physics called Matrix Product State models, and finally discrete directed acyclic graphical models.Chapter 1 explores the statistical problems of model selection and parameter identifiability from the perspective of algebraic geometry, in the case of hidden Markov models (HMMs) where all the hidden random variables are binary. Its main contributions are (1) a new parametrization for every such HMM via a birational map with an explicit inverse for recovering the hidden parameters in terms of observables, (2) a semialgebraic model membership test to determine if a discrete probability distribution can arise from such an HMM, and (3) minimal defining equations for the set of probability distributions arising from the 4-node fully binary model, com- prising 21 quadrics and 29 cubics, which were computed using Grobner bases in the cumulant coordinates of Bernd Sturmfels and Piotr Zwiernik. The new model parameters in (1) are rationally identifiable in the sense of Seth Sullivant, Luis David Garcia-Puente, and Sarah Spielvogel, and each model's Zariski closure is therefore a rational projective variety of dimension 5. Grobner basis computations for the model and its graph are found to be considerably faster using these parameters. In the case of two hidden states, (2) supersedes a previous algorithm of Alexander Schonhuth which is only generically defined, and the defining equations (3) yield new invariants for HMMs of all lengths ≥ 4. Such invariants have been used successfully in model selection problems in phylogenetics, and one can hope for similar applications in the case of HMMs.In Chapter 2, we study the representational power of matrix product states (MPS) with binary virtual bonds for entangled qubit systems. We do this by giving polynomial expressions in a pure quantum state's amplitudes which hold if and only if the state is a translation invariant matrix product state or a limit of such states. For systems with few qubits, we give these equations explicitly, considering both periodic and open boundary conditions. Using the classical theory of trace varieties and trace algebras, we explain the relationship between MPS and hidden Markov models and exploit this relationship to derive useful parameterizations of MPS. We present four conjectures on the identifiability of MPS parameters.Chapter 3 develops new parameters for use with directed acyclic graphical (DAG) models on discrete variables, which can simplify symbolic computations for tree models with hidden variables having more than two states. This development is the first step toward generalizing work of Smith and Zwiernik on binary trees, and makes it possible for some of the techniques used in Chapters 1 and 2 to be applied to graphical models with variables having more than two states." @default.
• W1830759839 created "2016-06-24" @default.
• W1830759839 creator A5034938018 @default.
• W1830759839 date "2013-01-01" @default.
• W1830759839 modified "2023-09-27" @default.
• W1830759839 title "Algebraic Geometry of Hidden Markov and Related Models" @default.
• W1830759839 cites W1515585009 @default.
• W1830759839 cites W1572665974 @default.
• W1830759839 cites W1584154100 @default.
• W1830759839 cites W1585991502 @default.
• W1830759839 cites W1853145084 @default.
• W1830759839 cites W1973309660 @default.
• W1830759839 cites W1974520228 @default.
• W1830759839 cites W1988171432 @default.
• W1830759839 cites W1991804994 @default.
• W1830759839 cites W2007321142 @default.
• W1830759839 cites W2016258127 @default.
• W1830759839 cites W2017995540 @default.
• W1830759839 cites W2034541289 @default.
• W1830759839 cites W2037181643 @default.
• W1830759839 cites W2037417588 @default.
• W1830759839 cites W2058801366 @default.
• W1830759839 cites W2063196329 @default.
• W1830759839 cites W2066972040 @default.
• W1830759839 cites W2077514957 @default.
• W1830759839 cites W2102892751 @default.
• W1830759839 cites W2105077826 @default.
• W1830759839 cites W2163929346 @default.
• W1830759839 cites W2168359188 @default.
• W1830759839 cites W2171090263 @default.
• W1830759839 cites W2292751606 @default.
• W1830759839 cites W2319466955 @default.
• W1830759839 cites W2804037619 @default.
• W1830759839 cites W2963069217 @default.
• W1830759839 cites W2963491532 @default.
• W1830759839 cites W2964269355 @default.
• W1830759839 cites W3096550277 @default.
• W1830759839 hasPublicationYear "2013" @default.
• W1830759839 type Work @default.
• W1830759839 sameAs 1830759839 @default.
• W1830759839 citedByCount "5" @default.
• W1830759839 countsByYear W18307598392014 @default.
• W1830759839 countsByYear W18307598392015 @default.
• W1830759839 countsByYear W18307598392016 @default.
• W1830759839 countsByYear W18307598392018 @default.
• W1830759839 crossrefType "journal-article" @default.
• W1830759839 hasAuthorship W1830759839A5034938018 @default.
• W1830759839 hasConcept C105795698 @default.
• W1830759839 hasConcept C11413529 @default.
• W1830759839 hasConcept C118615104 @default.
• W1830759839 hasConcept C121332964 @default.
• W1830759839 hasConcept C122770356 @default.
• W1830759839 hasConcept C136119220 @default.
• W1830759839 hasConcept C137002209 @default.
• W1830759839 hasConcept C154945302 @default.
• W1830759839 hasConcept C198082693 @default.
• W1830759839 hasConcept C202444582 @default.
• W1830759839 hasConcept C23224414 @default.
• W1830759839 hasConcept C33923547 @default.
• W1830759839 hasConcept C41008148 @default.
• W1830759839 hasConcept C45374587 @default.
• W1830759839 hasConcept C62520636 @default.
• W1830759839 hasConcept C68363185 @default.
• W1830759839 hasConcept C84114770 @default.
• W1830759839 hasConceptScore W1830759839C105795698 @default.
• W1830759839 hasConceptScore W1830759839C11413529 @default.
• W1830759839 hasConceptScore W1830759839C118615104 @default.
• W1830759839 hasConceptScore W1830759839C121332964 @default.
• W1830759839 hasConceptScore W1830759839C122770356 @default.
• W1830759839 hasConceptScore W1830759839C136119220 @default.
• W1830759839 hasConceptScore W1830759839C137002209 @default.
• W1830759839 hasConceptScore W1830759839C154945302 @default.
• W1830759839 hasConceptScore W1830759839C198082693 @default.
• W1830759839 hasConceptScore W1830759839C202444582 @default.
• W1830759839 hasConceptScore W1830759839C23224414 @default.
• W1830759839 hasConceptScore W1830759839C33923547 @default.
• W1830759839 hasConceptScore W1830759839C41008148 @default.
• W1830759839 hasConceptScore W1830759839C45374587 @default.
• W1830759839 hasConceptScore W1830759839C62520636 @default.
• W1830759839 hasConceptScore W1830759839C68363185 @default.
• W1830759839 hasConceptScore W1830759839C84114770 @default.
• W1830759839 hasLocation W18307598391 @default.
• W1830759839 hasOpenAccess W1830759839 @default.
• W1830759839 hasPrimaryLocation W18307598391 @default.
• W1830759839 hasRelatedWork W1246381107 @default.
• W1830759839 hasRelatedWork W1488779284 @default.
• W1830759839 hasRelatedWork W1530272429 @default.
• W1830759839 hasRelatedWork W1542359973 @default.
• W1830759839 hasRelatedWork W1554681636 @default.
• W1830759839 hasRelatedWork W1636040588 @default.
• W1830759839 hasRelatedWork W1692926957 @default.
• W1830759839 hasRelatedWork W1825959699 @default.
• W1830759839 hasRelatedWork W1982725637 @default.
• W1830759839 hasRelatedWork W2013912476 @default.
• W1830759839 hasRelatedWork W2024165284 @default.
• W1830759839 hasRelatedWork W2031216664 @default.
• W1830759839 hasRelatedWork W2031628971 @default.
• W1830759839 hasRelatedWork W2042901969 @default.
• W1830759839 hasRelatedWork W2069641044 @default.
• W1830759839 hasRelatedWork W2092332806 @default.

• next