Partial optimality by pruning for MAP-inference with general graphical models
We consider the energy minimization problem for undirected graphical models, also known as MAP-inference problem for Markov random fields which is NP-hard in general. We propose a novel polynomial time algorithm to obtain a part of its optimal non-relaxed integral solution. Our algorithm is initiali...
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
2016
|
| In: |
IEEE transactions on pattern analysis and machine intelligence
Year: 2015, Jahrgang: 38, Heft: 7, Pages: 1370-1382 |
| ISSN: | 1939-3539 |
| DOI: | 10.1109/TPAMI.2015.2484327 |
| Online-Zugang: | Verlag, lizenzpflichtig, Volltext: https://doi.org/10.1109/TPAMI.2015.2484327
|
| Verfasserangaben: | Paul Swoboda, Alexander Shekhovtsov, Jörg Hendrik Kappes, Christoph Schnörr, and Bogdan Savchynskyy |
MARC
| LEADER | 00000caa a22000002c 4500 | ||
|---|---|---|---|
| 001 | 1698831986 | ||
| 003 | DE-627 | ||
| 005 | 20220818100451.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 200526r20162015xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.1109/TPAMI.2015.2484327 |2 doi | |
| 035 | |a (DE-627)1698831986 | ||
| 035 | |a (DE-599)KXP1698831986 | ||
| 035 | |a (OCoLC)1341326559 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 27 |2 sdnb | ||
| 100 | 1 | |a Swoboda, Paul |e VerfasserIn |0 (DE-588)1066353379 |0 (DE-627)817351434 |0 (DE-576)425790231 |4 aut | |
| 245 | 1 | 0 | |a Partial optimality by pruning for MAP-inference with general graphical models |c Paul Swoboda, Alexander Shekhovtsov, Jörg Hendrik Kappes, Christoph Schnörr, and Bogdan Savchynskyy |
| 264 | 1 | |c 2016 | |
| 300 | |a 13 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Date of publication 12 Oct 2015 | ||
| 500 | |a Gesehen am 26.05.2020 | ||
| 520 | |a We consider the energy minimization problem for undirected graphical models, also known as MAP-inference problem for Markov random fields which is NP-hard in general. We propose a novel polynomial time algorithm to obtain a part of its optimal non-relaxed integral solution. Our algorithm is initialized with variables taking integral values in the solution of a convex relaxation of the MAP-inference problem and iteratively prunes those, which do not satisfy our criterion for partial optimality. We show that our pruning strategy is in a certain sense theoretically optimal. Also empirically our method outperforms previous approaches in terms of the number of persistently labelled variables. The method is very general, as it is applicable to models with arbitrary factors of an arbitrary order and can employ any solver for the considered relaxed problem. Our method's runtime is determined by the runtime of the convex relaxation solver for the MAP-inference problem. | ||
| 534 | |c 2015 | ||
| 650 | 4 | |a computational complexity | |
| 650 | 4 | |a convex relaxation solver | |
| 650 | 4 | |a directed graphs | |
| 650 | 4 | |a energy minimization | |
| 650 | 4 | |a energy minimization problem | |
| 650 | 4 | |a general graphical models | |
| 650 | 4 | |a Graphical models | |
| 650 | 4 | |a inference mechanisms | |
| 650 | 4 | |a Labeling | |
| 650 | 4 | |a local polytope | |
| 650 | 4 | |a MAP-inference | |
| 650 | 4 | |a MAP-inference problem | |
| 650 | 4 | |a Markov processes | |
| 650 | 4 | |a Markov random fields | |
| 650 | 4 | |a minimisation | |
| 650 | 4 | |a Minimization | |
| 650 | 4 | |a NP-hard problem | |
| 650 | 4 | |a optimal nonrelaxed integral solution | |
| 650 | 4 | |a partial optimality | |
| 650 | 4 | |a persistency | |
| 650 | 4 | |a polynomial time algorithm | |
| 650 | 4 | |a Polynomials | |
| 650 | 4 | |a pruning strategy | |
| 650 | 4 | |a Runtime | |
| 650 | 4 | |a Signal processing algorithms | |
| 650 | 4 | |a undirected graphical models | |
| 700 | 1 | |a Kappes, Jörg Hendrik |e VerfasserIn |0 (DE-588)1011935481 |0 (DE-627)660877821 |0 (DE-576)344648818 |4 aut | |
| 700 | 1 | |a Schnörr, Christoph |e VerfasserIn |0 (DE-588)1023033348 |0 (DE-627)717351017 |0 (DE-576)168404540 |4 aut | |
| 700 | 1 | |a Savchynskyy, Bogdan |e VerfasserIn |0 (DE-588)1066503605 |0 (DE-627)817659439 |0 (DE-576)426018737 |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |a Institute of Electrical and Electronics Engineers |t IEEE transactions on pattern analysis and machine intelligence |d New York, NY : IEEE, 1979 |g 38(2016), 7, Seite 1370-1382 |h Online-Ressource |w (DE-627)324486421 |w (DE-600)2027336-8 |w (DE-576)094110980 |x 1939-3539 |7 nnas |
| 773 | 1 | 8 | |g volume:38 |g year:2016 |g number:7 |g pages:1370-1382 |g extent:13 |a Partial optimality by pruning for MAP-inference with general graphical models |
| 856 | 4 | 0 | |u https://doi.org/10.1109/TPAMI.2015.2484327 |x Verlag |x Resolving-System |z lizenzpflichtig |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20200526 | ||
| 993 | |a Article | ||
| 994 | |a 2016 | ||
| 998 | |g 1066503605 |a Savchynskyy, Bogdan |m 1066503605:Savchynskyy, Bogdan |d 700000 |d 708000 |e 700000PS1066503605 |e 708000PS1066503605 |k 0/700000/ |k 1/700000/708000/ |p 5 |y j | ||
| 998 | |g 1023033348 |a Schnörr, Christoph |m 1023033348:Schnörr, Christoph |d 700000 |d 720000 |e 700000PS1023033348 |e 720000PS1023033348 |k 0/700000/ |k 1/700000/720000/ |p 4 | ||
| 998 | |g 1011935481 |a Kappes, Jörg Hendrik |m 1011935481:Kappes, Jörg Hendrik |d 700000 |d 708070 |e 700000PK1011935481 |e 708070PK1011935481 |k 0/700000/ |k 1/700000/708070/ |p 3 | ||
| 998 | |g 1066353379 |a Swoboda, Paul |m 1066353379:Swoboda, Paul |d 700000 |d 708070 |e 700000PS1066353379 |e 708070PS1066353379 |k 0/700000/ |k 1/700000/708070/ |p 1 |x j | ||
| 999 | |a KXP-PPN1698831986 |e 3671293527 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"person":[{"display":"Swoboda, Paul","family":"Swoboda","role":"aut","given":"Paul"},{"role":"aut","given":"Jörg Hendrik","display":"Kappes, Jörg Hendrik","family":"Kappes"},{"family":"Schnörr","display":"Schnörr, Christoph","role":"aut","given":"Christoph"},{"family":"Savchynskyy","display":"Savchynskyy, Bogdan","given":"Bogdan","role":"aut"}],"language":["eng"],"note":["Date of publication 12 Oct 2015","Gesehen am 26.05.2020"],"type":{"media":"Online-Ressource","bibl":"article-journal"},"origin":[{"dateIssuedDisp":"2016","dateIssuedKey":"2016"}],"title":[{"title_sort":"Partial optimality by pruning for MAP-inference with general graphical models","title":"Partial optimality by pruning for MAP-inference with general graphical models"}],"recId":"1698831986","name":{"displayForm":["Paul Swoboda, Alexander Shekhovtsov, Jörg Hendrik Kappes, Christoph Schnörr, and Bogdan Savchynskyy"]},"physDesc":[{"extent":"13 S."}],"relHost":[{"language":["eng"],"title":[{"subtitle":"TPAMI","title_sort":"IEEE transactions on pattern analysis and machine intelligence","title":"IEEE transactions on pattern analysis and machine intelligence"}],"note":["Gesehen am 07. März 2019"],"origin":[{"dateIssuedKey":"1979","publisher":"IEEE","dateIssuedDisp":"1979-","publisherPlace":"New York, NY"}],"titleAlt":[{"title":"Transactions on pattern analysis and machine intelligence"},{"title":"TPAMI"}],"type":{"media":"Online-Ressource","bibl":"periodical"},"part":{"text":"38(2016), 7, Seite 1370-1382","pages":"1370-1382","year":"2016","volume":"38","extent":"13","issue":"7"},"corporate":[{"role":"aut","display":"Institute of Electrical and Electronics Engineers"}],"disp":"Institute of Electrical and Electronics EngineersIEEE transactions on pattern analysis and machine intelligence","id":{"issn":["1939-3539"],"eki":["324486421"],"zdb":["2027336-8"]},"pubHistory":["1.1979 -"],"recId":"324486421","name":{"displayForm":["Institute of Electrical and Electronics Engineers"]},"physDesc":[{"extent":"Online-Ressource"}]}],"id":{"eki":["1698831986"],"doi":["10.1109/TPAMI.2015.2484327"]}} | ||
| SRT | |a SWOBODAPAUPARTIALOPT2016 | ||