DRiLLS: Deep Reinforcement Learning for Logic Synthesis

Logic synthesis requires extensive tuning of the synthesis optimization flow where the quality of results (QoR) depends on the sequence of optimizations used. Efficient design space exploration is challenging due to the exponential number of possible optimization permutations. Therefore, automating...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Proceedings of the ASP-DAC ... Asia and South Pacific Design Automation Conference S. 581 - 586
Hauptverfasser: Hosny, Abdelrahman, Hashemi, Soheil, Shalan, Mohamed, Reda, Sherief
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 01.01.2020
Schlagworte:
ISSN:2153-697X
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Abstract Logic synthesis requires extensive tuning of the synthesis optimization flow where the quality of results (QoR) depends on the sequence of optimizations used. Efficient design space exploration is challenging due to the exponential number of possible optimization permutations. Therefore, automating the optimization process is necessary. In this work, we propose a novel reinforcement learning-based methodology that navigates the optimization space without human intervention. We demonstrate the training of an Advantage Actor Critic (A2C) agent that seeks to minimize area subject to a timing constraint. Using the proposed methodology, designs can be optimized autonomously with no-humans in-loop. Evaluation on the comprehensive EPFL benchmark suite shows that the agent outperforms existing exploration methodologies and improves QoRs by an average of 13%.
AbstractList Logic synthesis requires extensive tuning of the synthesis optimization flow where the quality of results (QoR) depends on the sequence of optimizations used. Efficient design space exploration is challenging due to the exponential number of possible optimization permutations. Therefore, automating the optimization process is necessary. In this work, we propose a novel reinforcement learning-based methodology that navigates the optimization space without human intervention. We demonstrate the training of an Advantage Actor Critic (A2C) agent that seeks to minimize area subject to a timing constraint. Using the proposed methodology, designs can be optimized autonomously with no-humans in-loop. Evaluation on the comprehensive EPFL benchmark suite shows that the agent outperforms existing exploration methodologies and improves QoRs by an average of 13%.
Author Hashemi, Soheil
Shalan, Mohamed
Hosny, Abdelrahman
Reda, Sherief
Author_xml – sequence: 1
  givenname: Abdelrahman
  surname: Hosny
  fullname: Hosny, Abdelrahman
  organization: Brown University Providence,Computer Science Dept,RI
– sequence: 2
  givenname: Soheil
  surname: Hashemi
  fullname: Hashemi, Soheil
  organization: Brown University Providence,School of Engineering,RI
– sequence: 3
  givenname: Mohamed
  surname: Shalan
  fullname: Shalan, Mohamed
  organization: American University in Cairo,Computer Science Dept,Cairo,Egypt
– sequence: 4
  givenname: Sherief
  surname: Reda
  fullname: Reda, Sherief
  organization: Brown University Providence,School of Engineering,RI
BookMark eNotj81Kw0AURkdRsK19AjfxAVLvnZnbmXFX2voDA0qj4K5Mkps6YiclyaZvb8F-mwNnceAbi6vUJhbiHmGGCO5hUbznq8VSG0PzmQQJMweaiNyFGKORFjVKBZdiJJFUPnfm60ZM-_4HTiOQBmEkzGoTvS8esxXzIdtwTE3bVbznNGSeQ5di2mUnlfl2F6usOKbhm_vY34rrJvz2PD1zIj6f1h_Ll9y_Pb8uFz4PEmHIS3IgK8JyThYU1Q5IGhuQa83aNjWVilywlitErVSjaitVJeuguaFGspqIu_9uZObtoYv70B2355_qD8cLSJk
ContentType Conference Proceeding
DBID 6IE
6IL
CBEJK
RIE
RIL
DOI 10.1109/ASP-DAC47756.2020.9045559
DatabaseName IEEE Electronic Library (IEL) Conference Proceedings
IEEE Proceedings Order Plan All Online (POP All Online) 1998-present by volume
IEEE Xplore All Conference Proceedings
IEEE Electronic Library (IEL)
IEEE Proceedings Order Plans (POP All) 1998-Present
DatabaseTitleList
Database_xml – sequence: 1
  dbid: RIE
  name: IEEE Electronic Library (IEL)
  url: https://ieeexplore.ieee.org/
  sourceTypes: Publisher
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
Computer Science
EISBN 1728141230
9781728141237
EISSN 2153-697X
EndPage 586
ExternalDocumentID 9045559
Genre orig-research
GroupedDBID 5VS
6IE
6IF
6IL
6IN
AAWTH
ABLEC
ACGFS
ADZIZ
ALMA_UNASSIGNED_HOLDINGS
APO
AVWKF
BEFXN
BFFAM
BGNUA
BKEBE
BPEOZ
CBEJK
CHZPO
I07
IEGSK
M43
OCL
RIE
RIL
ID FETCH-LOGICAL-a210t-b5902c51b658035d905278a1ed4e48fd5b359a88ec11433f3d823c2da4ef5f2e3
IEDL.DBID RIE
IngestDate Wed Aug 27 05:55:38 EDT 2025
IsPeerReviewed false
IsScholarly false
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-a210t-b5902c51b658035d905278a1ed4e48fd5b359a88ec11433f3d823c2da4ef5f2e3
PageCount 6
ParticipantIDs ieee_primary_9045559
PublicationCentury 2000
PublicationDate 2020-Jan.
PublicationDateYYYYMMDD 2020-01-01
PublicationDate_xml – month: 01
  year: 2020
  text: 2020-Jan.
PublicationDecade 2020
PublicationTitle Proceedings of the ASP-DAC ... Asia and South Pacific Design Automation Conference
PublicationTitleAbbrev ASP-DAC
PublicationYear 2020
Publisher IEEE
Publisher_xml – name: IEEE
SSID ssj0000502710
ssib055574204
Score 2.151234
Snippet Logic synthesis requires extensive tuning of the synthesis optimization flow where the quality of results (QoR) depends on the sequence of optimizations used....
SourceID ieee
SourceType Publisher
StartPage 581
SubjectTerms Benchmark testing
Circuit synthesis
Delays
Optimization
Reinforcement learning
Space exploration
Tuning
Title DRiLLS: Deep Reinforcement Learning for Logic Synthesis
URI https://ieeexplore.ieee.org/document/9045559
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NTwIxEJ0AMUYvKGD8Tk08urC7bWnXGwGJB0IIqOFGuu2s4bIQPkz897a7C2rixVszh6aZtpn3pp03APccNVU8Fp7WTHksaXNP2SjuCaMsOGW6HWbPBW8DMRzK6TQaleBhXwuDiNnnM2y6YfaWbxZ661JlrcjiD4uAy1AWop3Xau3OjrVbkldQi1zX2xKuwD-Eu0JWs9WZjLxep8uE4O5vQug3i_l-NVbJ4kq_-r8VnUDju0CPjPah5xRKmNaguuvQQIoLW4PjH3KDdRC98XwwmDySHuKSjDETTdVZfpAUOqvvxJqIa8CsyeQztehwPV834LX_9NJ99orGCZ6yDG7jxU6TRfMgtvDCp9xEPg-FVAEahkwmhseUR0pK1JYNUZpQI0OqQ6MYJjwJkZ5BJV2keA5EUqNloIKgHRrH9WJECzokmpgJ5UfyAurOK7Nlro0xKxxy-bf5Co6c4_MUxjVUNqst3sCB_tjM16vbbEO_AMVOnY4
linkProvider IEEE
linkToHtml http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LTwIxEJ4gGh8XFDC-rYlHV3a3Ldv1RkCCcSUE0HAj3XbWcFkIDxP_vd0HqIkXb800aZo-Mt837XwDcMtRUclDz1KKSYtFdW5J48UtT0sDTpmqu-lzwVvgdbtiNPJ7Bbjb5MIgYvr5DO-TZvqWr6dqlYTKar7BHwYBb8E2Z8y1s2yt9ekxPYbm5eQiU_Y2lMuxd-EmF9asNQY9q9VoMs_jye8E177PR_xVWiX1LO3S_-Z0CNXvFD3S2zifIyhgXIbSukYDya9sGQ5-CA5WwGv1J0EweCAtxBnpYyqbqtIIIcmVVt-JMZGkBLMig8_Y4MPFZFGF1_bjsNmx8tIJljQcbmmFiSqL4k5oAIZNufZt7npCOqgZMhFpHlLuSyFQGT5EaUS1cKlytWQY8chFegzFeBrjCRBBtRKOdJy6qxO2FyIa2CFQh8yTti9OoZKsyniWqWOM8wU5-9t8DXud4UswDp66z-ewn2xCFtC4gOJyvsJL2FEfy8lifpVu7hcsGKDV
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=proceeding&rft.title=Proceedings+of+the+ASP-DAC+...+Asia+and+South+Pacific+Design+Automation+Conference&rft.atitle=DRiLLS%3A+Deep+Reinforcement+Learning+for+Logic+Synthesis&rft.au=Hosny%2C+Abdelrahman&rft.au=Hashemi%2C+Soheil&rft.au=Shalan%2C+Mohamed&rft.au=Reda%2C+Sherief&rft.date=2020-01-01&rft.pub=IEEE&rft.eissn=2153-697X&rft.spage=581&rft.epage=586&rft_id=info:doi/10.1109%2FASP-DAC47756.2020.9045559&rft.externalDocID=9045559