On proportional volume sampling for experimental design in general spaces
Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d....
Ausführliche Beschreibung
Gespeichert in:
| Autor*in: |
Poinas, Arnaud [verfasserIn] |
|---|
| Format: |
Artikel |
|---|---|
| Sprache: |
Englisch |
| Erschienen: |
2022 |
|---|
| Schlagwörter: |
|---|
| Anmerkung: |
© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022 |
|---|
| Übergeordnetes Werk: |
Enthalten in: Statistics and computing - Springer US, 1991, 33(2022), 1 vom: 31. Dez. |
|---|---|
| Übergeordnetes Werk: |
volume:33 ; year:2022 ; number:1 ; day:31 ; month:12 |
| Links: |
|---|
| DOI / URN: |
10.1007/s11222-022-10115-0 |
|---|
| Katalog-ID: |
OLC2080207318 |
|---|
| LEADER | 01000caa a22002652 4500 | ||
|---|---|---|---|
| 001 | OLC2080207318 | ||
| 003 | DE-627 | ||
| 005 | 20230506155144.0 | ||
| 007 | tu | ||
| 008 | 230131s2022 xx ||||| 00| ||eng c | ||
| 024 | 7 | |a 10.1007/s11222-022-10115-0 |2 doi | |
| 035 | |a (DE-627)OLC2080207318 | ||
| 035 | |a (DE-He213)s11222-022-10115-0-p | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rakwb | ||
| 041 | |a eng | ||
| 082 | 0 | 4 | |a 004 |a 620 |q VZ |
| 100 | 1 | |a Poinas, Arnaud |e verfasserin |0 (orcid)0000-0002-3553-6695 |4 aut | |
| 245 | 1 | 0 | |a On proportional volume sampling for experimental design in general spaces |
| 264 | 1 | |c 2022 | |
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a ohne Hilfsmittel zu benutzen |b n |2 rdamedia | ||
| 338 | |a Band |b nc |2 rdacarrier | ||
| 500 | |a © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022 | ||
| 520 | |a Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d. sampling. PVS strikes the balance between design nodes that jointly fill the design space, while marginally staying in regions of high mass under the solution of a relaxed convex version of the original problem. In this paper, we examine some of the statistical implications of a new variant of PVS for (possibly Bayesian) optimal design. Using point process machinery, we treat the case of a generic Polish design space. We show that not only are known A-optimality approximation guarantees preserved, but we obtain similar guarantees for D-optimal design that tighten recent results. Moreover, we show that our PVS variant can be sampled in polynomial time. Unfortunately, in spite of its elegance and tractability, we demonstrate on a simple example that the practical implications of general PVS are likely limited. In the second part of the paper, we focus on applications and investigate the use of PVS as a subroutine for stochastic search heuristics. We demonstrate that PVS is a robust addition to the practitioner’s toolbox, especially when the regression functions are nonstandard and the design space, while low-dimensional, has a complicated shape (e.g., nonlinear boundaries, several connected components). | ||
| 650 | 4 | |a Bayesian optimal design | |
| 650 | 4 | |a Volume sampling | |
| 650 | 4 | |a Determinantal point processes | |
| 650 | 4 | |a Search heuristics | |
| 700 | 1 | |a Bardenet, Rémi |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Statistics and computing |d Springer US, 1991 |g 33(2022), 1 vom: 31. Dez. |w (DE-627)131007963 |w (DE-600)1087487-2 |w (DE-576)052732894 |x 0960-3174 |7 nnns |
| 773 | 1 | 8 | |g volume:33 |g year:2022 |g number:1 |g day:31 |g month:12 |
| 856 | 4 | 1 | |u https://doi.org/10.1007/s11222-022-10115-0 |z lizenzpflichtig |3 Volltext |
| 912 | |a GBV_USEFLAG_A | ||
| 912 | |a SYSFLAG_A | ||
| 912 | |a GBV_OLC | ||
| 912 | |a SSG-OLC-TEC | ||
| 912 | |a SSG-OLC-MAT | ||
| 951 | |a AR | ||
| 952 | |d 33 |j 2022 |e 1 |b 31 |c 12 | ||
| author_variant |
a p ap r b rb |
|---|---|
| matchkey_str |
article:09603174:2022----::nrprinloueapigoeprmnadsg |
| hierarchy_sort_str |
2022 |
| publishDate |
2022 |
| allfields |
10.1007/s11222-022-10115-0 doi (DE-627)OLC2080207318 (DE-He213)s11222-022-10115-0-p DE-627 ger DE-627 rakwb eng 004 620 VZ Poinas, Arnaud verfasserin (orcid)0000-0002-3553-6695 aut On proportional volume sampling for experimental design in general spaces 2022 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022 Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d. sampling. PVS strikes the balance between design nodes that jointly fill the design space, while marginally staying in regions of high mass under the solution of a relaxed convex version of the original problem. In this paper, we examine some of the statistical implications of a new variant of PVS for (possibly Bayesian) optimal design. Using point process machinery, we treat the case of a generic Polish design space. We show that not only are known A-optimality approximation guarantees preserved, but we obtain similar guarantees for D-optimal design that tighten recent results. Moreover, we show that our PVS variant can be sampled in polynomial time. Unfortunately, in spite of its elegance and tractability, we demonstrate on a simple example that the practical implications of general PVS are likely limited. In the second part of the paper, we focus on applications and investigate the use of PVS as a subroutine for stochastic search heuristics. We demonstrate that PVS is a robust addition to the practitioner’s toolbox, especially when the regression functions are nonstandard and the design space, while low-dimensional, has a complicated shape (e.g., nonlinear boundaries, several connected components). Bayesian optimal design Volume sampling Determinantal point processes Search heuristics Bardenet, Rémi aut Enthalten in Statistics and computing Springer US, 1991 33(2022), 1 vom: 31. Dez. (DE-627)131007963 (DE-600)1087487-2 (DE-576)052732894 0960-3174 nnns volume:33 year:2022 number:1 day:31 month:12 https://doi.org/10.1007/s11222-022-10115-0 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT AR 33 2022 1 31 12 |
| spelling |
10.1007/s11222-022-10115-0 doi (DE-627)OLC2080207318 (DE-He213)s11222-022-10115-0-p DE-627 ger DE-627 rakwb eng 004 620 VZ Poinas, Arnaud verfasserin (orcid)0000-0002-3553-6695 aut On proportional volume sampling for experimental design in general spaces 2022 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022 Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d. sampling. PVS strikes the balance between design nodes that jointly fill the design space, while marginally staying in regions of high mass under the solution of a relaxed convex version of the original problem. In this paper, we examine some of the statistical implications of a new variant of PVS for (possibly Bayesian) optimal design. Using point process machinery, we treat the case of a generic Polish design space. We show that not only are known A-optimality approximation guarantees preserved, but we obtain similar guarantees for D-optimal design that tighten recent results. Moreover, we show that our PVS variant can be sampled in polynomial time. Unfortunately, in spite of its elegance and tractability, we demonstrate on a simple example that the practical implications of general PVS are likely limited. In the second part of the paper, we focus on applications and investigate the use of PVS as a subroutine for stochastic search heuristics. We demonstrate that PVS is a robust addition to the practitioner’s toolbox, especially when the regression functions are nonstandard and the design space, while low-dimensional, has a complicated shape (e.g., nonlinear boundaries, several connected components). Bayesian optimal design Volume sampling Determinantal point processes Search heuristics Bardenet, Rémi aut Enthalten in Statistics and computing Springer US, 1991 33(2022), 1 vom: 31. Dez. (DE-627)131007963 (DE-600)1087487-2 (DE-576)052732894 0960-3174 nnns volume:33 year:2022 number:1 day:31 month:12 https://doi.org/10.1007/s11222-022-10115-0 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT AR 33 2022 1 31 12 |
| allfields_unstemmed |
10.1007/s11222-022-10115-0 doi (DE-627)OLC2080207318 (DE-He213)s11222-022-10115-0-p DE-627 ger DE-627 rakwb eng 004 620 VZ Poinas, Arnaud verfasserin (orcid)0000-0002-3553-6695 aut On proportional volume sampling for experimental design in general spaces 2022 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022 Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d. sampling. PVS strikes the balance between design nodes that jointly fill the design space, while marginally staying in regions of high mass under the solution of a relaxed convex version of the original problem. In this paper, we examine some of the statistical implications of a new variant of PVS for (possibly Bayesian) optimal design. Using point process machinery, we treat the case of a generic Polish design space. We show that not only are known A-optimality approximation guarantees preserved, but we obtain similar guarantees for D-optimal design that tighten recent results. Moreover, we show that our PVS variant can be sampled in polynomial time. Unfortunately, in spite of its elegance and tractability, we demonstrate on a simple example that the practical implications of general PVS are likely limited. In the second part of the paper, we focus on applications and investigate the use of PVS as a subroutine for stochastic search heuristics. We demonstrate that PVS is a robust addition to the practitioner’s toolbox, especially when the regression functions are nonstandard and the design space, while low-dimensional, has a complicated shape (e.g., nonlinear boundaries, several connected components). Bayesian optimal design Volume sampling Determinantal point processes Search heuristics Bardenet, Rémi aut Enthalten in Statistics and computing Springer US, 1991 33(2022), 1 vom: 31. Dez. (DE-627)131007963 (DE-600)1087487-2 (DE-576)052732894 0960-3174 nnns volume:33 year:2022 number:1 day:31 month:12 https://doi.org/10.1007/s11222-022-10115-0 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT AR 33 2022 1 31 12 |
| allfieldsGer |
10.1007/s11222-022-10115-0 doi (DE-627)OLC2080207318 (DE-He213)s11222-022-10115-0-p DE-627 ger DE-627 rakwb eng 004 620 VZ Poinas, Arnaud verfasserin (orcid)0000-0002-3553-6695 aut On proportional volume sampling for experimental design in general spaces 2022 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022 Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d. sampling. PVS strikes the balance between design nodes that jointly fill the design space, while marginally staying in regions of high mass under the solution of a relaxed convex version of the original problem. In this paper, we examine some of the statistical implications of a new variant of PVS for (possibly Bayesian) optimal design. Using point process machinery, we treat the case of a generic Polish design space. We show that not only are known A-optimality approximation guarantees preserved, but we obtain similar guarantees for D-optimal design that tighten recent results. Moreover, we show that our PVS variant can be sampled in polynomial time. Unfortunately, in spite of its elegance and tractability, we demonstrate on a simple example that the practical implications of general PVS are likely limited. In the second part of the paper, we focus on applications and investigate the use of PVS as a subroutine for stochastic search heuristics. We demonstrate that PVS is a robust addition to the practitioner’s toolbox, especially when the regression functions are nonstandard and the design space, while low-dimensional, has a complicated shape (e.g., nonlinear boundaries, several connected components). Bayesian optimal design Volume sampling Determinantal point processes Search heuristics Bardenet, Rémi aut Enthalten in Statistics and computing Springer US, 1991 33(2022), 1 vom: 31. Dez. (DE-627)131007963 (DE-600)1087487-2 (DE-576)052732894 0960-3174 nnns volume:33 year:2022 number:1 day:31 month:12 https://doi.org/10.1007/s11222-022-10115-0 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT AR 33 2022 1 31 12 |
| allfieldsSound |
10.1007/s11222-022-10115-0 doi (DE-627)OLC2080207318 (DE-He213)s11222-022-10115-0-p DE-627 ger DE-627 rakwb eng 004 620 VZ Poinas, Arnaud verfasserin (orcid)0000-0002-3553-6695 aut On proportional volume sampling for experimental design in general spaces 2022 Text txt rdacontent ohne Hilfsmittel zu benutzen n rdamedia Band nc rdacarrier © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022 Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d. sampling. PVS strikes the balance between design nodes that jointly fill the design space, while marginally staying in regions of high mass under the solution of a relaxed convex version of the original problem. In this paper, we examine some of the statistical implications of a new variant of PVS for (possibly Bayesian) optimal design. Using point process machinery, we treat the case of a generic Polish design space. We show that not only are known A-optimality approximation guarantees preserved, but we obtain similar guarantees for D-optimal design that tighten recent results. Moreover, we show that our PVS variant can be sampled in polynomial time. Unfortunately, in spite of its elegance and tractability, we demonstrate on a simple example that the practical implications of general PVS are likely limited. In the second part of the paper, we focus on applications and investigate the use of PVS as a subroutine for stochastic search heuristics. We demonstrate that PVS is a robust addition to the practitioner’s toolbox, especially when the regression functions are nonstandard and the design space, while low-dimensional, has a complicated shape (e.g., nonlinear boundaries, several connected components). Bayesian optimal design Volume sampling Determinantal point processes Search heuristics Bardenet, Rémi aut Enthalten in Statistics and computing Springer US, 1991 33(2022), 1 vom: 31. Dez. (DE-627)131007963 (DE-600)1087487-2 (DE-576)052732894 0960-3174 nnns volume:33 year:2022 number:1 day:31 month:12 https://doi.org/10.1007/s11222-022-10115-0 lizenzpflichtig Volltext GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT AR 33 2022 1 31 12 |
| language |
English |
| source |
Enthalten in Statistics and computing 33(2022), 1 vom: 31. Dez. volume:33 year:2022 number:1 day:31 month:12 |
| sourceStr |
Enthalten in Statistics and computing 33(2022), 1 vom: 31. Dez. volume:33 year:2022 number:1 day:31 month:12 |
| format_phy_str_mv |
Article |
| institution |
findex.gbv.de |
| topic_facet |
Bayesian optimal design Volume sampling Determinantal point processes Search heuristics |
| dewey-raw |
004 |
| isfreeaccess_bool |
false |
| container_title |
Statistics and computing |
| authorswithroles_txt_mv |
Poinas, Arnaud @@aut@@ Bardenet, Rémi @@aut@@ |
| publishDateDaySort_date |
2022-12-31T00:00:00Z |
| hierarchy_top_id |
131007963 |
| dewey-sort |
14 |
| id |
OLC2080207318 |
| language_de |
englisch |
| fullrecord |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">OLC2080207318</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230506155144.0</controlfield><controlfield tag="007">tu</controlfield><controlfield tag="008">230131s2022 xx ||||| 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1007/s11222-022-10115-0</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)OLC2080207318</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-He213)s11222-022-10115-0-p</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">004</subfield><subfield code="a">620</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Poinas, Arnaud</subfield><subfield code="e">verfasserin</subfield><subfield code="0">(orcid)0000-0002-3553-6695</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">On proportional volume sampling for experimental design in general spaces</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2022</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">ohne Hilfsmittel zu benutzen</subfield><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Band</subfield><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="500" ind1=" " ind2=" "><subfield code="a">© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d. sampling. PVS strikes the balance between design nodes that jointly fill the design space, while marginally staying in regions of high mass under the solution of a relaxed convex version of the original problem. In this paper, we examine some of the statistical implications of a new variant of PVS for (possibly Bayesian) optimal design. Using point process machinery, we treat the case of a generic Polish design space. We show that not only are known A-optimality approximation guarantees preserved, but we obtain similar guarantees for D-optimal design that tighten recent results. Moreover, we show that our PVS variant can be sampled in polynomial time. Unfortunately, in spite of its elegance and tractability, we demonstrate on a simple example that the practical implications of general PVS are likely limited. In the second part of the paper, we focus on applications and investigate the use of PVS as a subroutine for stochastic search heuristics. We demonstrate that PVS is a robust addition to the practitioner’s toolbox, especially when the regression functions are nonstandard and the design space, while low-dimensional, has a complicated shape (e.g., nonlinear boundaries, several connected components).</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Bayesian optimal design</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Volume sampling</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Determinantal point processes</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Search heuristics</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Bardenet, Rémi</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">Statistics and computing</subfield><subfield code="d">Springer US, 1991</subfield><subfield code="g">33(2022), 1 vom: 31. Dez.</subfield><subfield code="w">(DE-627)131007963</subfield><subfield code="w">(DE-600)1087487-2</subfield><subfield code="w">(DE-576)052732894</subfield><subfield code="x">0960-3174</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:33</subfield><subfield code="g">year:2022</subfield><subfield code="g">number:1</subfield><subfield code="g">day:31</subfield><subfield code="g">month:12</subfield></datafield><datafield tag="856" ind1="4" ind2="1"><subfield code="u">https://doi.org/10.1007/s11222-022-10115-0</subfield><subfield code="z">lizenzpflichtig</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_OLC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-TEC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-MAT</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">33</subfield><subfield code="j">2022</subfield><subfield code="e">1</subfield><subfield code="b">31</subfield><subfield code="c">12</subfield></datafield></record></collection>
|
| author |
Poinas, Arnaud |
| spellingShingle |
Poinas, Arnaud ddc 004 misc Bayesian optimal design misc Volume sampling misc Determinantal point processes misc Search heuristics On proportional volume sampling for experimental design in general spaces |
| authorStr |
Poinas, Arnaud |
| ppnlink_with_tag_str_mv |
@@773@@(DE-627)131007963 |
| format |
Article |
| dewey-ones |
004 - Data processing & computer science 620 - Engineering & allied operations |
| delete_txt_mv |
keep |
| author_role |
aut aut |
| collection |
OLC |
| remote_str |
false |
| illustrated |
Not Illustrated |
| issn |
0960-3174 |
| topic_title |
004 620 VZ On proportional volume sampling for experimental design in general spaces Bayesian optimal design Volume sampling Determinantal point processes Search heuristics |
| topic |
ddc 004 misc Bayesian optimal design misc Volume sampling misc Determinantal point processes misc Search heuristics |
| topic_unstemmed |
ddc 004 misc Bayesian optimal design misc Volume sampling misc Determinantal point processes misc Search heuristics |
| topic_browse |
ddc 004 misc Bayesian optimal design misc Volume sampling misc Determinantal point processes misc Search heuristics |
| format_facet |
Aufsätze Gedruckte Aufsätze |
| format_main_str_mv |
Text Zeitschrift/Artikel |
| carriertype_str_mv |
nc |
| hierarchy_parent_title |
Statistics and computing |
| hierarchy_parent_id |
131007963 |
| dewey-tens |
000 - Computer science, knowledge & systems 620 - Engineering |
| hierarchy_top_title |
Statistics and computing |
| isfreeaccess_txt |
false |
| familylinks_str_mv |
(DE-627)131007963 (DE-600)1087487-2 (DE-576)052732894 |
| title |
On proportional volume sampling for experimental design in general spaces |
| ctrlnum |
(DE-627)OLC2080207318 (DE-He213)s11222-022-10115-0-p |
| title_full |
On proportional volume sampling for experimental design in general spaces |
| author_sort |
Poinas, Arnaud |
| journal |
Statistics and computing |
| journalStr |
Statistics and computing |
| lang_code |
eng |
| isOA_bool |
false |
| dewey-hundreds |
000 - Computer science, information & general works 600 - Technology |
| recordtype |
marc |
| publishDateSort |
2022 |
| contenttype_str_mv |
txt |
| author_browse |
Poinas, Arnaud Bardenet, Rémi |
| container_volume |
33 |
| class |
004 620 VZ |
| format_se |
Aufsätze |
| author-letter |
Poinas, Arnaud |
| doi_str_mv |
10.1007/s11222-022-10115-0 |
| normlink |
(ORCID)0000-0002-3553-6695 |
| normlink_prefix_str_mv |
(orcid)0000-0002-3553-6695 |
| dewey-full |
004 620 |
| title_sort |
on proportional volume sampling for experimental design in general spaces |
| title_auth |
On proportional volume sampling for experimental design in general spaces |
| abstract |
Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d. sampling. PVS strikes the balance between design nodes that jointly fill the design space, while marginally staying in regions of high mass under the solution of a relaxed convex version of the original problem. In this paper, we examine some of the statistical implications of a new variant of PVS for (possibly Bayesian) optimal design. Using point process machinery, we treat the case of a generic Polish design space. We show that not only are known A-optimality approximation guarantees preserved, but we obtain similar guarantees for D-optimal design that tighten recent results. Moreover, we show that our PVS variant can be sampled in polynomial time. Unfortunately, in spite of its elegance and tractability, we demonstrate on a simple example that the practical implications of general PVS are likely limited. In the second part of the paper, we focus on applications and investigate the use of PVS as a subroutine for stochastic search heuristics. We demonstrate that PVS is a robust addition to the practitioner’s toolbox, especially when the regression functions are nonstandard and the design space, while low-dimensional, has a complicated shape (e.g., nonlinear boundaries, several connected components). © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022 |
| abstractGer |
Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d. sampling. PVS strikes the balance between design nodes that jointly fill the design space, while marginally staying in regions of high mass under the solution of a relaxed convex version of the original problem. In this paper, we examine some of the statistical implications of a new variant of PVS for (possibly Bayesian) optimal design. Using point process machinery, we treat the case of a generic Polish design space. We show that not only are known A-optimality approximation guarantees preserved, but we obtain similar guarantees for D-optimal design that tighten recent results. Moreover, we show that our PVS variant can be sampled in polynomial time. Unfortunately, in spite of its elegance and tractability, we demonstrate on a simple example that the practical implications of general PVS are likely limited. In the second part of the paper, we focus on applications and investigate the use of PVS as a subroutine for stochastic search heuristics. We demonstrate that PVS is a robust addition to the practitioner’s toolbox, especially when the regression functions are nonstandard and the design space, while low-dimensional, has a complicated shape (e.g., nonlinear boundaries, several connected components). © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022 |
| abstract_unstemmed |
Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d. sampling. PVS strikes the balance between design nodes that jointly fill the design space, while marginally staying in regions of high mass under the solution of a relaxed convex version of the original problem. In this paper, we examine some of the statistical implications of a new variant of PVS for (possibly Bayesian) optimal design. Using point process machinery, we treat the case of a generic Polish design space. We show that not only are known A-optimality approximation guarantees preserved, but we obtain similar guarantees for D-optimal design that tighten recent results. Moreover, we show that our PVS variant can be sampled in polynomial time. Unfortunately, in spite of its elegance and tractability, we demonstrate on a simple example that the practical implications of general PVS are likely limited. In the second part of the paper, we focus on applications and investigate the use of PVS as a subroutine for stochastic search heuristics. We demonstrate that PVS is a robust addition to the practitioner’s toolbox, especially when the regression functions are nonstandard and the design space, while low-dimensional, has a complicated shape (e.g., nonlinear boundaries, several connected components). © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022 |
| collection_details |
GBV_USEFLAG_A SYSFLAG_A GBV_OLC SSG-OLC-TEC SSG-OLC-MAT |
| container_issue |
1 |
| title_short |
On proportional volume sampling for experimental design in general spaces |
| url |
https://doi.org/10.1007/s11222-022-10115-0 |
| remote_bool |
false |
| author2 |
Bardenet, Rémi |
| author2Str |
Bardenet, Rémi |
| ppnlink |
131007963 |
| mediatype_str_mv |
n |
| isOA_txt |
false |
| hochschulschrift_bool |
false |
| doi_str |
10.1007/s11222-022-10115-0 |
| up_date |
2024-07-04T03:12:50.951Z |
| _version_ |
1803616540125298688 |
| fullrecord_marcxml |
<?xml version="1.0" encoding="UTF-8"?><collection xmlns="http://www.loc.gov/MARC21/slim"><record><leader>01000caa a22002652 4500</leader><controlfield tag="001">OLC2080207318</controlfield><controlfield tag="003">DE-627</controlfield><controlfield tag="005">20230506155144.0</controlfield><controlfield tag="007">tu</controlfield><controlfield tag="008">230131s2022 xx ||||| 00| ||eng c</controlfield><datafield tag="024" ind1="7" ind2=" "><subfield code="a">10.1007/s11222-022-10115-0</subfield><subfield code="2">doi</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-627)OLC2080207318</subfield></datafield><datafield tag="035" ind1=" " ind2=" "><subfield code="a">(DE-He213)s11222-022-10115-0-p</subfield></datafield><datafield tag="040" ind1=" " ind2=" "><subfield code="a">DE-627</subfield><subfield code="b">ger</subfield><subfield code="c">DE-627</subfield><subfield code="e">rakwb</subfield></datafield><datafield tag="041" ind1=" " ind2=" "><subfield code="a">eng</subfield></datafield><datafield tag="082" ind1="0" ind2="4"><subfield code="a">004</subfield><subfield code="a">620</subfield><subfield code="q">VZ</subfield></datafield><datafield tag="100" ind1="1" ind2=" "><subfield code="a">Poinas, Arnaud</subfield><subfield code="e">verfasserin</subfield><subfield code="0">(orcid)0000-0002-3553-6695</subfield><subfield code="4">aut</subfield></datafield><datafield tag="245" ind1="1" ind2="0"><subfield code="a">On proportional volume sampling for experimental design in general spaces</subfield></datafield><datafield tag="264" ind1=" " ind2="1"><subfield code="c">2022</subfield></datafield><datafield tag="336" ind1=" " ind2=" "><subfield code="a">Text</subfield><subfield code="b">txt</subfield><subfield code="2">rdacontent</subfield></datafield><datafield tag="337" ind1=" " ind2=" "><subfield code="a">ohne Hilfsmittel zu benutzen</subfield><subfield code="b">n</subfield><subfield code="2">rdamedia</subfield></datafield><datafield tag="338" ind1=" " ind2=" "><subfield code="a">Band</subfield><subfield code="b">nc</subfield><subfield code="2">rdacarrier</subfield></datafield><datafield tag="500" ind1=" " ind2=" "><subfield code="a">© The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2022</subfield></datafield><datafield tag="520" ind1=" " ind2=" "><subfield code="a">Abstract Optimal design for linear regression is a fundamental task in statistics. For finite design spaces, recent progress has shown that random designs drawn using proportional volume sampling (PVS for short) lead to polynomial-time algorithms with approximation guarantees that outperform i.i.d. sampling. PVS strikes the balance between design nodes that jointly fill the design space, while marginally staying in regions of high mass under the solution of a relaxed convex version of the original problem. In this paper, we examine some of the statistical implications of a new variant of PVS for (possibly Bayesian) optimal design. Using point process machinery, we treat the case of a generic Polish design space. We show that not only are known A-optimality approximation guarantees preserved, but we obtain similar guarantees for D-optimal design that tighten recent results. Moreover, we show that our PVS variant can be sampled in polynomial time. Unfortunately, in spite of its elegance and tractability, we demonstrate on a simple example that the practical implications of general PVS are likely limited. In the second part of the paper, we focus on applications and investigate the use of PVS as a subroutine for stochastic search heuristics. We demonstrate that PVS is a robust addition to the practitioner’s toolbox, especially when the regression functions are nonstandard and the design space, while low-dimensional, has a complicated shape (e.g., nonlinear boundaries, several connected components).</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Bayesian optimal design</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Volume sampling</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Determinantal point processes</subfield></datafield><datafield tag="650" ind1=" " ind2="4"><subfield code="a">Search heuristics</subfield></datafield><datafield tag="700" ind1="1" ind2=" "><subfield code="a">Bardenet, Rémi</subfield><subfield code="4">aut</subfield></datafield><datafield tag="773" ind1="0" ind2="8"><subfield code="i">Enthalten in</subfield><subfield code="t">Statistics and computing</subfield><subfield code="d">Springer US, 1991</subfield><subfield code="g">33(2022), 1 vom: 31. Dez.</subfield><subfield code="w">(DE-627)131007963</subfield><subfield code="w">(DE-600)1087487-2</subfield><subfield code="w">(DE-576)052732894</subfield><subfield code="x">0960-3174</subfield><subfield code="7">nnns</subfield></datafield><datafield tag="773" ind1="1" ind2="8"><subfield code="g">volume:33</subfield><subfield code="g">year:2022</subfield><subfield code="g">number:1</subfield><subfield code="g">day:31</subfield><subfield code="g">month:12</subfield></datafield><datafield tag="856" ind1="4" ind2="1"><subfield code="u">https://doi.org/10.1007/s11222-022-10115-0</subfield><subfield code="z">lizenzpflichtig</subfield><subfield code="3">Volltext</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_USEFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SYSFLAG_A</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">GBV_OLC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-TEC</subfield></datafield><datafield tag="912" ind1=" " ind2=" "><subfield code="a">SSG-OLC-MAT</subfield></datafield><datafield tag="951" ind1=" " ind2=" "><subfield code="a">AR</subfield></datafield><datafield tag="952" ind1=" " ind2=" "><subfield code="d">33</subfield><subfield code="j">2022</subfield><subfield code="e">1</subfield><subfield code="b">31</subfield><subfield code="c">12</subfield></datafield></record></collection>
|
| score |
7.4008036 |