Scalable Neural Decoder for Topological Surface Codes

With the advent of noisy intermediate-scale quantum (NISQ) devices, practical quantum computing has seemingly come into reach. However, to go beyond proof-of-principle calculations, the current processing architectures will need to scale up to larger quantum circuits which will require fast and scal...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:Physical review letters Ročník 128; číslo 8; s. 080505
Hlavní autori: Meinerz, Kai, Park, Chae-Yeun, Trebst, Simon
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: United States 25.02.2022
Predmet:
ISSN:0031-9007, 1079-7114, 1079-7114
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Abstract With the advent of noisy intermediate-scale quantum (NISQ) devices, practical quantum computing has seemingly come into reach. However, to go beyond proof-of-principle calculations, the current processing architectures will need to scale up to larger quantum circuits which will require fast and scalable algorithms for quantum error correction. Here, we present a neural network based decoder that, for a family of stabilizer codes subject to depolarizing noise and syndrome measurement errors, is scalable to tens of thousands of qubits (in contrast to other recent machine learning inspired decoders) and exhibits faster decoding times than the state-of-the-art union find decoder for a wide range of error rates (down to 1%). The key innovation is to autodecode error syndromes on small scales by shifting a preprocessing window over the underlying code, akin to a convolutional neural network in pattern recognition approaches. We show that such a preprocessing step allows to effectively reduce the error rate by up to 2 orders of magnitude in practical applications and, by detecting correlation effects, shifts the actual error threshold up to fifteen percent higher than the threshold of conventional error correction algorithms such as union find or minimum weight perfect matching, even in the presence of measurement errors. An in situ implementation of such a machine learning-assisted quantum error correction will be a decisive step to push the entanglement frontier beyond the NISQ horizon.
AbstractList With the advent of noisy intermediate-scale quantum (NISQ) devices, practical quantum computing has seemingly come into reach. However, to go beyond proof-of-principle calculations, the current processing architectures will need to scale up to larger quantum circuits which will require fast and scalable algorithms for quantum error correction. Here, we present a neural network based decoder that, for a family of stabilizer codes subject to depolarizing noise and syndrome measurement errors, is scalable to tens of thousands of qubits (in contrast to other recent machine learning inspired decoders) and exhibits faster decoding times than the state-of-the-art union find decoder for a wide range of error rates (down to 1%). The key innovation is to autodecode error syndromes on small scales by shifting a preprocessing window over the underlying code, akin to a convolutional neural network in pattern recognition approaches. We show that such a preprocessing step allows to effectively reduce the error rate by up to 2 orders of magnitude in practical applications and, by detecting correlation effects, shifts the actual error threshold up to fifteen percent higher than the threshold of conventional error correction algorithms such as union find or minimum weight perfect matching, even in the presence of measurement errors. An in situ implementation of such a machine learning-assisted quantum error correction will be a decisive step to push the entanglement frontier beyond the NISQ horizon.
With the advent of noisy intermediate-scale quantum (NISQ) devices, practical quantum computing has seemingly come into reach. However, to go beyond proof-of-principle calculations, the current processing architectures will need to scale up to larger quantum circuits which will require fast and scalable algorithms for quantum error correction. Here, we present a neural network based decoder that, for a family of stabilizer codes subject to depolarizing noise and syndrome measurement errors, is scalable to tens of thousands of qubits (in contrast to other recent machine learning inspired decoders) and exhibits faster decoding times than the state-of-the-art union find decoder for a wide range of error rates (down to 1%). The key innovation is to autodecode error syndromes on small scales by shifting a preprocessing window over the underlying code, akin to a convolutional neural network in pattern recognition approaches. We show that such a preprocessing step allows to effectively reduce the error rate by up to 2 orders of magnitude in practical applications and, by detecting correlation effects, shifts the actual error threshold up to fifteen percent higher than the threshold of conventional error correction algorithms such as union find or minimum weight perfect matching, even in the presence of measurement errors. An in situ implementation of such a machine learning-assisted quantum error correction will be a decisive step to push the entanglement frontier beyond the NISQ horizon.With the advent of noisy intermediate-scale quantum (NISQ) devices, practical quantum computing has seemingly come into reach. However, to go beyond proof-of-principle calculations, the current processing architectures will need to scale up to larger quantum circuits which will require fast and scalable algorithms for quantum error correction. Here, we present a neural network based decoder that, for a family of stabilizer codes subject to depolarizing noise and syndrome measurement errors, is scalable to tens of thousands of qubits (in contrast to other recent machine learning inspired decoders) and exhibits faster decoding times than the state-of-the-art union find decoder for a wide range of error rates (down to 1%). The key innovation is to autodecode error syndromes on small scales by shifting a preprocessing window over the underlying code, akin to a convolutional neural network in pattern recognition approaches. We show that such a preprocessing step allows to effectively reduce the error rate by up to 2 orders of magnitude in practical applications and, by detecting correlation effects, shifts the actual error threshold up to fifteen percent higher than the threshold of conventional error correction algorithms such as union find or minimum weight perfect matching, even in the presence of measurement errors. An in situ implementation of such a machine learning-assisted quantum error correction will be a decisive step to push the entanglement frontier beyond the NISQ horizon.
ArticleNumber 080505
Author Park, Chae-Yeun
Trebst, Simon
Meinerz, Kai
Author_xml – sequence: 1
  givenname: Kai
  orcidid: 0000-0002-9141-7113
  surname: Meinerz
  fullname: Meinerz, Kai
– sequence: 2
  givenname: Chae-Yeun
  surname: Park
  fullname: Park, Chae-Yeun
– sequence: 3
  givenname: Simon
  orcidid: 0000-0002-1479-9736
  surname: Trebst
  fullname: Trebst, Simon
BackLink https://www.ncbi.nlm.nih.gov/pubmed/35275669$$D View this record in MEDLINE/PubMed
BookMark eNqFkMtOwzAQRS1URB_wC1WWbFI8cfyIxAaVp1QBomVtOe4EgtK62AlS_x6jthJiw2oWc-4dzRmS3tqtkZAx0AkAZRfP79vwgl8zbNsJZGpCFeWUH5EBUFmkEiDvkQGlDNKCUtknwxA-KKWQCXVC-oxnkgtRDAifW9OYssHkETtvmuQarVuiTyrnk4XbuMa91RFJ5p2vjMVkGrfhlBxXpgl4tp8j8np7s5jep7Onu4fp1Sy1DKBNDYgSDCtRCgVQVVRxFFygzVFxi2wJwI2JBOVFxbjIreEYaSlZrrAs2Yic73o33n12GFq9qoPFpjFrdF3QmWBKAhOqiOh4j3blCpd64-uV8Vt9eDUClzvAeheCx0rbujVt7datN3Wjgeofs_qXWR3N6p3ZGBd_4ocL_wS_Aas9f_s
CitedBy_id crossref_primary_10_1103_PhysRevResearch_6_013154
crossref_primary_10_1016_j_engappai_2025_110808
crossref_primary_10_1038_s41586_024_08148_8
crossref_primary_10_1088_2058_9565_adc3ba
crossref_primary_10_1038_s41467_023_42482_1
crossref_primary_10_1002_qute_202500158
crossref_primary_10_1103_PhysRevX_13_031007
crossref_primary_10_1088_2058_9565_ace64d
crossref_primary_10_1145_3505637
crossref_primary_10_1007_s11128_024_04377_y
crossref_primary_10_1103_PhysRevApplied_19_034050
crossref_primary_10_1103_PhysRevA_111_012419
crossref_primary_10_1038_s42005_024_01883_4
crossref_primary_10_1007_s11128_025_04826_2
crossref_primary_10_1103_PhysRevResearch_7_023181
crossref_primary_10_1088_2399_1984_aceba6
crossref_primary_10_1103_PRXQuantum_6_010360
crossref_primary_10_1038_s41534_025_01033_w
crossref_primary_10_1103_PRXQuantum_4_040344
crossref_primary_10_1145_3733239
crossref_primary_10_1103_PhysRevResearch_7_013029
crossref_primary_10_1007_s11128_023_03898_2
crossref_primary_10_1088_1674_1056_adab63
crossref_primary_10_1103_PhysRevResearch_6_L032004
Cites_doi 10.1103/PhysRevA.96.062302
10.1103/PhysRevLett.108.180501
10.1103/PhysRevX.2.021004
10.22331/q-2019-09-02-183
10.22331/q-2020-08-24-310
10.1103/PhysRevA.102.042411
10.22331/q-2021-12-02-595
10.1103/PhysRevA.76.042319
10.1088/1367-2630/aaf29e
10.1016/j.physleta.2020.126353
10.1038/s41586-019-1666-5
10.1103/PhysRevLett.122.200501
10.1063/1.1499754
10.1088/2632-2153/abc609
10.22331/q-2018-08-06-79
10.1103/PhysRevX.11.031039
10.1090/psapm/068/2762145
10.26421/QIC14.9-10-1
10.1103/PhysRevLett.104.050504
10.1038/s41567-020-0920-y
10.1103/PhysRevA.102.032411
10.4153/CJM-1965-045-4
10.1103/PhysRevResearch.2.033042
10.1088/2058-9565/aad1f7
10.1103/PhysRevLett.119.030501
10.1103/PhysRevResearch.2.023230
10.1016/S0003-4916(02)00018-0
10.1088/2058-9565/aa955a
10.1103/PhysRevA.86.032324
10.1038/s41598-017-11266-1
ContentType Journal Article
DBID AAYXX
CITATION
CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1103/PhysRevLett.128.080505
DatabaseName CrossRef
Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle CrossRef
MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList MEDLINE
MEDLINE - Academic
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod fulltext_linktorsrc
Discipline Physics
EISSN 1079-7114
ExternalDocumentID 35275669
10_1103_PhysRevLett_128_080505
Genre Journal Article
GroupedDBID ---
-DZ
-~X
123
186
2-P
29O
3MX
3O-
41~
5VS
6TJ
85S
8NH
8WZ
9M8
A6W
AAYJJ
AAYXX
ABSSX
ABUFD
ACBEA
ACGFO
ACKIV
ACNCT
ADXHL
AECSF
AENEX
AEQTI
AETEA
AFFNX
AFGMR
AGDNE
AJQPL
ALMA_UNASSIGNED_HOLDINGS
APKKM
AUAIK
CITATION
CS3
D0L
DU5
EBS
EJD
ER.
F5P
H~9
MVM
N9A
NEJ
NHB
NPBMV
OHT
OK1
P0-
P2P
RNS
ROL
S7W
SJN
T9H
TN5
UBC
UBE
VOH
WH7
XOL
XSW
YNT
YYP
ZCG
ZPR
ZY4
~02
CGR
CUY
CVF
ECM
EIF
NPM
UCJ
VQA
7X8
ID FETCH-LOGICAL-c311t-a16b1a3be76811ff085e656ec4e85ce3d115aa1a3059f3564ca5ebe777348ebb3
IEDL.DBID 3MX
ISICitedReferencesCount 35
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000767187300003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0031-9007
1079-7114
IngestDate Thu Oct 02 11:03:17 EDT 2025
Thu Jan 02 22:54:58 EST 2025
Tue Nov 18 22:18:31 EST 2025
Sat Nov 29 05:56:13 EST 2025
IsPeerReviewed true
IsScholarly true
Issue 8
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c311t-a16b1a3be76811ff085e656ec4e85ce3d115aa1a3059f3564ca5ebe777348ebb3
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ORCID 0000-0002-9141-7113
0000-0002-1479-9736
PMID 35275669
PQID 2638713689
PQPubID 23479
ParticipantIDs proquest_miscellaneous_2638713689
pubmed_primary_35275669
crossref_citationtrail_10_1103_PhysRevLett_128_080505
crossref_primary_10_1103_PhysRevLett_128_080505
PublicationCentury 2000
PublicationDate 2022-02-25
PublicationDateYYYYMMDD 2022-02-25
PublicationDate_xml – month: 02
  year: 2022
  text: 2022-02-25
  day: 25
PublicationDecade 2020
PublicationPlace United States
PublicationPlace_xml – name: United States
PublicationTitle Physical review letters
PublicationTitleAlternate Phys Rev Lett
PublicationYear 2022
References G. Duclos-Cianci (PhysRevLett.128.080505Cc16R1) 2010
PhysRevLett.128.080505Cc28R1
PhysRevLett.128.080505Cc29R1
PhysRevLett.128.080505Cc26R1
PhysRevLett.128.080505Cc27R1
PhysRevLett.128.080505Cc9R1
PhysRevLett.128.080505Cc7R1
PhysRevLett.128.080505Cc6R1
N P. Jouppi (PhysRevLett.128.080505Cc36R1) 2017
PhysRevLett.128.080505Cc24R1
PhysRevLett.128.080505Cc4R1
PhysRevLett.128.080505Cc25R1
PhysRevLett.128.080505Cc46R1
PhysRevLett.128.080505Cc3R1
PhysRevLett.128.080505Cc22R1
PhysRevLett.128.080505Cc2R1
PhysRevLett.128.080505Cc23R1
PhysRevLett.128.080505Cc44R1
PhysRevLett.128.080505Cc1R1
PhysRevLett.128.080505Cc20R1
PhysRevLett.128.080505Cc21R1
PhysRevLett.128.080505Cc40R1
PhysRevLett.128.080505Cc41R1
PhysRevLett.128.080505Cc15R1
PhysRevLett.128.080505Cc18R1
PhysRevLett.128.080505Cc17R1
E. L. Lawler (PhysRevLett.128.080505Cc39R1) 1976
PhysRevLett.128.080505Cc12R1
PhysRevLett.128.080505Cc13R1
PhysRevLett.128.080505Cc31R1
PhysRevLett.128.080505Cc32R1
PhysRevLett.128.080505Cc10R1
PhysRevLett.128.080505Cc30R1
References_xml – ident: PhysRevLett.128.080505Cc4R1
  doi: 10.1103/PhysRevA.96.062302
– ident: PhysRevLett.128.080505Cc40R1
  doi: 10.1103/PhysRevLett.108.180501
– ident: PhysRevLett.128.080505Cc46R1
  doi: 10.1103/PhysRevX.2.021004
– ident: PhysRevLett.128.080505Cc26R1
  doi: 10.22331/q-2019-09-02-183
– ident: PhysRevLett.128.080505Cc30R1
  doi: 10.22331/q-2020-08-24-310
– ident: PhysRevLett.128.080505Cc27R1
  doi: 10.1103/PhysRevA.102.042411
– ident: PhysRevLett.128.080505Cc18R1
  doi: 10.22331/q-2021-12-02-595
– ident: PhysRevLett.128.080505Cc3R1
  doi: 10.1103/PhysRevA.76.042319
– ident: PhysRevLett.128.080505Cc24R1
  doi: 10.1088/1367-2630/aaf29e
– ident: PhysRevLett.128.080505Cc29R1
  doi: 10.1016/j.physleta.2020.126353
– ident: PhysRevLett.128.080505Cc2R1
  doi: 10.1038/s41586-019-1666-5
– volume-title: 2010 IEEE Information Theory Workshop
  year: 2010
  ident: PhysRevLett.128.080505Cc16R1
– ident: PhysRevLett.128.080505Cc25R1
  doi: 10.1103/PhysRevLett.122.200501
– ident: PhysRevLett.128.080505Cc13R1
  doi: 10.1063/1.1499754
– ident: PhysRevLett.128.080505Cc32R1
  doi: 10.1088/2632-2153/abc609
– ident: PhysRevLett.128.080505Cc1R1
  doi: 10.22331/q-2018-08-06-79
– ident: PhysRevLett.128.080505Cc44R1
  doi: 10.1103/PhysRevX.11.031039
– volume-title: Combinatorial Optimization: Networks and Matroids
  year: 1976
  ident: PhysRevLett.128.080505Cc39R1
– ident: PhysRevLett.128.080505Cc6R1
  doi: 10.1090/psapm/068/2762145
– ident: PhysRevLett.128.080505Cc17R1
  doi: 10.26421/QIC14.9-10-1
– ident: PhysRevLett.128.080505Cc15R1
  doi: 10.1103/PhysRevLett.104.050504
– ident: PhysRevLett.128.080505Cc10R1
  doi: 10.1038/s41567-020-0920-y
– ident: PhysRevLett.128.080505Cc31R1
  doi: 10.1103/PhysRevA.102.032411
– ident: PhysRevLett.128.080505Cc12R1
  doi: 10.4153/CJM-1965-045-4
– ident: PhysRevLett.128.080505Cc41R1
  doi: 10.1103/PhysRevResearch.2.033042
– ident: PhysRevLett.128.080505Cc23R1
  doi: 10.1088/2058-9565/aad1f7
– volume-title: Proceedings of the 2017 ACM/IEEE 44th Annual International Symposium on Computer Architecture (ISCA)
  year: 2017
  ident: PhysRevLett.128.080505Cc36R1
– ident: PhysRevLett.128.080505Cc21R1
  doi: 10.1103/PhysRevLett.119.030501
– ident: PhysRevLett.128.080505Cc28R1
  doi: 10.1103/PhysRevResearch.2.023230
– ident: PhysRevLett.128.080505Cc7R1
  doi: 10.1016/S0003-4916(02)00018-0
– ident: PhysRevLett.128.080505Cc20R1
  doi: 10.1088/2058-9565/aa955a
– ident: PhysRevLett.128.080505Cc9R1
  doi: 10.1103/PhysRevA.86.032324
– ident: PhysRevLett.128.080505Cc22R1
  doi: 10.1038/s41598-017-11266-1
SSID ssj0001268
Score 2.6051292
Snippet With the advent of noisy intermediate-scale quantum (NISQ) devices, practical quantum computing has seemingly come into reach. However, to go beyond...
SourceID proquest
pubmed
crossref
SourceType Aggregation Database
Index Database
Enrichment Source
StartPage 080505
SubjectTerms Algorithms
Computing Methodologies
Machine Learning
Neural Networks, Computer
Quantum Theory
Title Scalable Neural Decoder for Topological Surface Codes
URI https://www.ncbi.nlm.nih.gov/pubmed/35275669
https://www.proquest.com/docview/2638713689
Volume 128
WOSCitedRecordID wos000767187300003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVABR
  databaseName: American Physical Society Journals
  customDbUrl:
  eissn: 1079-7114
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001268
  issn: 0031-9007
  databaseCode: 3MX
  dateStart: 20020101
  isFulltext: true
  titleUrlDefault: https://journals.aps.org/
  providerName: American Physical Society
– providerCode: PRVIAO
  databaseName: SCOAP3 Journals
  customDbUrl:
  eissn: 1079-7114
  dateEnd: 99991231
  omitProxy: false
  ssIdentifier: ssj0001268
  issn: 0031-9007
  databaseCode: ER.
  dateStart: 20180101
  isFulltext: true
  titleUrlDefault: https://scoap3.org/
  providerName: SCOAP3 (Sponsoring Consortium for Open Access Publishing in Particle Physics)
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlZ1LS8NAEICHUhS8-H7UFxG8ps1ms0l6lGrxoEVshdzCZjoLgqSStP39ziZpUbCIl1yyuwkzw-y37DwAbpGENyUybsa86wYk2Q-G2nMxRJoGyAyCVdeSp2g0ipOk_9IC7_cbfOHJno2EfKWlzW7psjPtMuKoumhpHNiWBfI5Wbte4Ye165U27sCLmpTgzcv83I02IGa11Qz3_v-T-7DbYKVzV9vBAbQoP4TtKrwTyyNQY1aFTZJybDEOHnhPNpe9cBhZnUndJ8FqyxkvCqORnAG_LY_hbfgwGTy6TcMEF6UQc1eLMBNaZsRnCCGMYZwi5jXCgGKFJKcseq15BDOVkSoMUCtWYhTZEjeUZfIE2vkspzNwTGyQtEGTCQoCE-qpH8UmFAaZiLRWHVArwaXYVBO3TS0-0upU4cn0m0hSFklai6QDvfW8z7qexp8zblZ6Sdn07X2Gzmm2KFOffQefscO434HTWmHrNZkrIybV_vm_v3cBO75NcbBp6-oS2vNiQVewhcv5e1lcV_bGzyiJvwBOctRe
linkProvider American Physical Society
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Scalable+Neural+Decoder+for+Topological+Surface+Codes&rft.jtitle=Physical+review+letters&rft.au=Meinerz%2C+Kai&rft.au=Park%2C+Chae-Yeun&rft.au=Trebst%2C+Simon&rft.date=2022-02-25&rft.eissn=1079-7114&rft.volume=128&rft.issue=8&rft.spage=080505&rft_id=info:doi/10.1103%2FPhysRevLett.128.080505&rft_id=info%3Apmid%2F35275669&rft.externalDocID=35275669
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0031-9007&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0031-9007&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0031-9007&client=summon