Generalized normalizing flows via Markov chains:
Normalizing flows, diffusion normalizing flows and variational autoencoders are powerful generative models. This Element provides a unified framework to handle these approaches via Markov chains. The authors consider stochastic normalizing flows as a pair of Markov chains fulfilling some properties,...
Gespeichert in:
Hauptverfasser: | , , |
---|---|
Format: | Elektronisch E-Book |
Sprache: | English |
Veröffentlicht: |
Cambridge
Cambridge University Press
2023
|
Schriftenreihe: | Cambridge elements
|
Schlagworte: | |
Online-Zugang: | BSB01 BTU01 FHN01 Volltext |
Zusammenfassung: | Normalizing flows, diffusion normalizing flows and variational autoencoders are powerful generative models. This Element provides a unified framework to handle these approaches via Markov chains. The authors consider stochastic normalizing flows as a pair of Markov chains fulfilling some properties, and show how many state-of-the-art models for data generation fit into this framework. Indeed numerical simulations show that including stochastic layers improves the expressivity of the network and allows for generating multimodal distributions from unimodal ones. The Markov chains point of view enables the coupling of both deterministic layers as invertible neural networks and stochastic layers as Metropolis-Hasting layers, Langevin layers, variational autoencoders and diffusion normalizing flows in a mathematically sound way. The authors' framework establishes a useful mathematical tool to combine the various approaches |
Beschreibung: | Previously issued in print: 2022. - Includes bibliographical references |
Beschreibung: | 1 Online-Ressource (57 Seiten) Illustrationen |
ISBN: | 9781009331012 |
DOI: | 10.1017/9781009331012 |
Internformat
MARC
LEADER | 00000nmm a2200000zc 4500 | ||
---|---|---|---|
001 | BV048823798 | ||
003 | DE-604 | ||
005 | 00000000000000.0 | ||
007 | cr|uuu---uuuuu | ||
008 | 230221s2023 |||| o||u| ||||||eng d | ||
020 | |a 9781009331012 |c Online |9 978-1-00-933101-2 | ||
024 | 7 | |a 10.1017/9781009331012 |2 doi | |
035 | |a (ZDB-20-CBO)CR9781009331012 | ||
035 | |a (OCoLC)1371326425 | ||
035 | |a (DE-599)BVBBV048823798 | ||
040 | |a DE-604 |b ger |e rda | ||
041 | 0 | |a eng | |
049 | |a DE-12 |a DE-92 |a DE-634 | ||
082 | 0 | |a 519.233 | |
100 | 1 | |a Hagemann, Paul |d 1997- |0 (DE-588)1281526576 |4 aut | |
245 | 1 | 0 | |a Generalized normalizing flows via Markov chains |c Paul Lyonel Hagemann, Johannes Hertrich, Gabriele Steidl |
264 | 1 | |a Cambridge |b Cambridge University Press |c 2023 | |
300 | |a 1 Online-Ressource (57 Seiten) |b Illustrationen | ||
336 | |b txt |2 rdacontent | ||
336 | |b sti |2 rdacontent | ||
337 | |b c |2 rdamedia | ||
338 | |b cr |2 rdacarrier | ||
490 | 0 | |a Cambridge elements | |
500 | |a Previously issued in print: 2022. - Includes bibliographical references | ||
520 | |a Normalizing flows, diffusion normalizing flows and variational autoencoders are powerful generative models. This Element provides a unified framework to handle these approaches via Markov chains. The authors consider stochastic normalizing flows as a pair of Markov chains fulfilling some properties, and show how many state-of-the-art models for data generation fit into this framework. Indeed numerical simulations show that including stochastic layers improves the expressivity of the network and allows for generating multimodal distributions from unimodal ones. The Markov chains point of view enables the coupling of both deterministic layers as invertible neural networks and stochastic layers as Metropolis-Hasting layers, Langevin layers, variational autoencoders and diffusion normalizing flows in a mathematically sound way. The authors' framework establishes a useful mathematical tool to combine the various approaches | ||
650 | 4 | |a Markov processes | |
700 | 1 | |a Hertrich, Johannes |d ca. 20./21. Jh. |0 (DE-588)1281526762 |4 aut | |
700 | 1 | |a Steidl, Gabriele |d 1963- |0 (DE-588)1067421165 |4 aut | |
776 | 0 | 8 | |i Erscheint auch als |n Druck-Ausgabe |z 978-1-00-933100-5 |
856 | 4 | 0 | |u https://doi.org/10.1017/9781009331012 |x Verlag |z URL des Erstveröffentlichers |3 Volltext |
912 | |a ZDB-20-CBO | ||
999 | |a oai:aleph.bib-bvb.de:BVB01-034089487 | ||
966 | e | |u https://doi.org/10.1017/9781009331012 |l BSB01 |p ZDB-20-CBO |q BSB_PDA_CBO |x Verlag |3 Volltext | |
966 | e | |u https://doi.org/10.1017/9781009331012 |l BTU01 |p ZDB-20-CBO |q BTU_PDA_CBO |x Verlag |3 Volltext | |
966 | e | |u https://doi.org/10.1017/9781009331012 |l FHN01 |p ZDB-20-CBO |q FHN_PDA_CBO |x Verlag |3 Volltext |
Datensatz im Suchindex
_version_ | 1804184918623780864 |
---|---|
adam_txt | |
any_adam_object | |
any_adam_object_boolean | |
author | Hagemann, Paul 1997- Hertrich, Johannes ca. 20./21. Jh Steidl, Gabriele 1963- |
author_GND | (DE-588)1281526576 (DE-588)1281526762 (DE-588)1067421165 |
author_facet | Hagemann, Paul 1997- Hertrich, Johannes ca. 20./21. Jh Steidl, Gabriele 1963- |
author_role | aut aut aut |
author_sort | Hagemann, Paul 1997- |
author_variant | p h ph j h jh g s gs |
building | Verbundindex |
bvnumber | BV048823798 |
collection | ZDB-20-CBO |
ctrlnum | (ZDB-20-CBO)CR9781009331012 (OCoLC)1371326425 (DE-599)BVBBV048823798 |
dewey-full | 519.233 |
dewey-hundreds | 500 - Natural sciences and mathematics |
dewey-ones | 519 - Probabilities and applied mathematics |
dewey-raw | 519.233 |
dewey-search | 519.233 |
dewey-sort | 3519.233 |
dewey-tens | 510 - Mathematics |
discipline | Mathematik |
discipline_str_mv | Mathematik |
doi_str_mv | 10.1017/9781009331012 |
format | Electronic eBook |
fullrecord | <?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>02750nmm a2200445zc 4500</leader><controlfield tag="001">BV048823798</controlfield><controlfield tag="003">DE-604</controlfield><controlfield tag="005">00000000000000.0</controlfield><controlfield tag="007">cr|uuu---uuuuu</controlfield><controlfield tag="008">230221s2023 |||| o||u| ||||||eng d</controlfield><datafield tag="020" ind1=" " ind2=" "><subfield code="a">9781009331012</subfield><subfield code="c">Online</subfield><subfield code="9">978-1-00-933101-2</subfield></datafield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1017/9781009331012</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(ZDB-20-CBO)CR9781009331012</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(OCoLC)1371326425</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-599)BVBBV048823798</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-604</subfield><subfield code="b">ger</subfield><subfield code="e">rda</subfield></datafield><datafield tag="041" ind1="0" ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="049" ind1=" " ind2=" "><subfield code="a">DE-12</subfield><subfield code="a">DE-92</subfield><subfield code="a">DE-634</subfield></datafield><datafield tag="082" ind1="0" ind2=" "><subfield code="a">519.233</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Hagemann, Paul</subfield><subfield code="d">1997-</subfield><subfield code="0">(DE-588)1281526576</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">Generalized normalizing flows via Markov chains</subfield><subfield code="c">Paul Lyonel Hagemann, Johannes Hertrich, Gabriele Steidl</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="a">Cambridge</subfield><subfield code="b">Cambridge University Press</subfield><subfield code="c">2023</subfield></datafield><datafield tag="300" ind1=" " ind2=" "><subfield code="a">1 Online-Ressource (57 Seiten)</subfield><subfield code="b">Illustrationen</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="b">sti</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="b">c</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="b">cr</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="490" ind1="0" ind2=" "><subfield code="a">Cambridge elements</subfield></datafield><datafield tag="500" ind1=" " ind2=" "><subfield code="a">Previously issued in print: 2022. - Includes bibliographical references</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Normalizing flows, diffusion normalizing flows and variational autoencoders are powerful generative models. This Element provides a unified framework to handle these approaches via Markov chains. The authors consider stochastic normalizing flows as a pair of Markov chains fulfilling some properties, and show how many state-of-the-art models for data generation fit into this framework. Indeed numerical simulations show that including stochastic layers improves the expressivity of the network and allows for generating multimodal distributions from unimodal ones. The Markov chains point of view enables the coupling of both deterministic layers as invertible neural networks and stochastic layers as Metropolis-Hasting layers, Langevin layers, variational autoencoders and diffusion normalizing flows in a mathematically sound way. The authors' framework establishes a useful mathematical tool to combine the various approaches</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Markov processes</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Hertrich, Johannes</subfield><subfield code="d">ca. 20./21. Jh.</subfield><subfield code="0">(DE-588)1281526762</subfield><subfield code="4">aut</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Steidl, Gabriele</subfield><subfield code="d">1963-</subfield><subfield code="0">(DE-588)1067421165</subfield><subfield code="4">aut</subfield></datafield><datafield tag="776" ind1="0" ind2="8"><subfield code="i">Erscheint auch als</subfield><subfield code="n">Druck-Ausgabe</subfield><subfield code="z">978-1-00-933100-5</subfield></datafield><datafield tag="856" ind1="4" ind2="0"><subfield code="u">https://doi.org/10.1017/9781009331012</subfield><subfield code="x">Verlag</subfield><subfield code="z">URL des Erstveröffentlichers</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">ZDB-20-CBO</subfield></datafield><datafield tag="999" ind1=" " ind2=" "><subfield code="a">oai:aleph.bib-bvb.de:BVB01-034089487</subfield></datafield><datafield tag="966" ind1="e" ind2=" "><subfield code="u">https://doi.org/10.1017/9781009331012</subfield><subfield code="l">BSB01</subfield><subfield code="p">ZDB-20-CBO</subfield><subfield code="q">BSB_PDA_CBO</subfield><subfield code="x">Verlag</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="966" ind1="e" ind2=" "><subfield code="u">https://doi.org/10.1017/9781009331012</subfield><subfield code="l">BTU01</subfield><subfield code="p">ZDB-20-CBO</subfield><subfield code="q">BTU_PDA_CBO</subfield><subfield code="x">Verlag</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="966" ind1="e" ind2=" "><subfield code="u">https://doi.org/10.1017/9781009331012</subfield><subfield code="l">FHN01</subfield><subfield code="p">ZDB-20-CBO</subfield><subfield code="q">FHN_PDA_CBO</subfield><subfield code="x">Verlag</subfield><subfield code="3">Volltext</subfield></datafield></record></collection> |
id | DE-604.BV048823798 |
illustrated | Not Illustrated |
index_date | 2024-07-03T21:33:54Z |
indexdate | 2024-07-10T09:46:58Z |
institution | BVB |
isbn | 9781009331012 |
language | English |
oai_aleph_id | oai:aleph.bib-bvb.de:BVB01-034089487 |
oclc_num | 1371326425 |
open_access_boolean | |
owner | DE-12 DE-92 DE-634 |
owner_facet | DE-12 DE-92 DE-634 |
physical | 1 Online-Ressource (57 Seiten) Illustrationen |
psigel | ZDB-20-CBO ZDB-20-CBO BSB_PDA_CBO ZDB-20-CBO BTU_PDA_CBO ZDB-20-CBO FHN_PDA_CBO |
publishDate | 2023 |
publishDateSearch | 2023 |
publishDateSort | 2023 |
publisher | Cambridge University Press |
record_format | marc |
series2 | Cambridge elements |
spelling | Hagemann, Paul 1997- (DE-588)1281526576 aut Generalized normalizing flows via Markov chains Paul Lyonel Hagemann, Johannes Hertrich, Gabriele Steidl Cambridge Cambridge University Press 2023 1 Online-Ressource (57 Seiten) Illustrationen txt rdacontent sti rdacontent c rdamedia cr rdacarrier Cambridge elements Previously issued in print: 2022. - Includes bibliographical references Normalizing flows, diffusion normalizing flows and variational autoencoders are powerful generative models. This Element provides a unified framework to handle these approaches via Markov chains. The authors consider stochastic normalizing flows as a pair of Markov chains fulfilling some properties, and show how many state-of-the-art models for data generation fit into this framework. Indeed numerical simulations show that including stochastic layers improves the expressivity of the network and allows for generating multimodal distributions from unimodal ones. The Markov chains point of view enables the coupling of both deterministic layers as invertible neural networks and stochastic layers as Metropolis-Hasting layers, Langevin layers, variational autoencoders and diffusion normalizing flows in a mathematically sound way. The authors' framework establishes a useful mathematical tool to combine the various approaches Markov processes Hertrich, Johannes ca. 20./21. Jh. (DE-588)1281526762 aut Steidl, Gabriele 1963- (DE-588)1067421165 aut Erscheint auch als Druck-Ausgabe 978-1-00-933100-5 https://doi.org/10.1017/9781009331012 Verlag URL des Erstveröffentlichers Volltext |
spellingShingle | Hagemann, Paul 1997- Hertrich, Johannes ca. 20./21. Jh Steidl, Gabriele 1963- Generalized normalizing flows via Markov chains Markov processes |
title | Generalized normalizing flows via Markov chains |
title_auth | Generalized normalizing flows via Markov chains |
title_exact_search | Generalized normalizing flows via Markov chains |
title_exact_search_txtP | Generalized normalizing flows via Markov chains |
title_full | Generalized normalizing flows via Markov chains Paul Lyonel Hagemann, Johannes Hertrich, Gabriele Steidl |
title_fullStr | Generalized normalizing flows via Markov chains Paul Lyonel Hagemann, Johannes Hertrich, Gabriele Steidl |
title_full_unstemmed | Generalized normalizing flows via Markov chains Paul Lyonel Hagemann, Johannes Hertrich, Gabriele Steidl |
title_short | Generalized normalizing flows via Markov chains |
title_sort | generalized normalizing flows via markov chains |
topic | Markov processes |
topic_facet | Markov processes |
url | https://doi.org/10.1017/9781009331012 |
work_keys_str_mv | AT hagemannpaul generalizednormalizingflowsviamarkovchains AT hertrichjohannes generalizednormalizingflowsviamarkovchains AT steidlgabriele generalizednormalizingflowsviamarkovchains |