Finding path and cycle counting formulae in graphs with Deep Reinforcement Learning
This paper presents Grammar Reinforcement Learning (GRL), a reinforcement learning algorithm that uses Monte Carlo Tree Search (MCTS) and a transformer architecture that models a Pushdown Automaton (PDA) within a context-free grammar (CFG) framework. Taking as use case the problem of efficiently cou...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Journal Article |
| Language: | English |
| Published: |
02-10-2024
|
| Subjects: | |
| Online Access: | Get full text |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Abstract | This paper presents Grammar Reinforcement Learning (GRL), a reinforcement
learning algorithm that uses Monte Carlo Tree Search (MCTS) and a transformer
architecture that models a Pushdown Automaton (PDA) within a context-free
grammar (CFG) framework. Taking as use case the problem of efficiently counting
paths and cycles in graphs, a key challenge in network analysis, computer
science, biology, and social sciences, GRL discovers new matrix-based formulas
for path/cycle counting that improve computational efficiency by factors of two
to six w.r.t state-of-the-art approaches. Our contributions include: (i) a
framework for generating gramformers that operate within a CFG, (ii) the
development of GRL for optimizing formulas within grammatical structures, and
(iii) the discovery of novel formulas for graph substructure counting, leading
to significant computational improvements. |
|---|---|
| AbstractList | This paper presents Grammar Reinforcement Learning (GRL), a reinforcement
learning algorithm that uses Monte Carlo Tree Search (MCTS) and a transformer
architecture that models a Pushdown Automaton (PDA) within a context-free
grammar (CFG) framework. Taking as use case the problem of efficiently counting
paths and cycles in graphs, a key challenge in network analysis, computer
science, biology, and social sciences, GRL discovers new matrix-based formulas
for path/cycle counting that improve computational efficiency by factors of two
to six w.r.t state-of-the-art approaches. Our contributions include: (i) a
framework for generating gramformers that operate within a CFG, (ii) the
development of GRL for optimizing formulas within grammatical structures, and
(iii) the discovery of novel formulas for graph substructure counting, leading
to significant computational improvements. |
| Author | Raveaux, Romain Piquenot, Jason Bérar, Maxime Ramel, Jean-Yves Adam, Sébastien Héroux, Pierre |
| Author_xml | – sequence: 1 givenname: Jason surname: Piquenot fullname: Piquenot, Jason – sequence: 2 givenname: Maxime surname: Bérar fullname: Bérar, Maxime – sequence: 3 givenname: Pierre surname: Héroux fullname: Héroux, Pierre – sequence: 4 givenname: Jean-Yves surname: Ramel fullname: Ramel, Jean-Yves – sequence: 5 givenname: Romain surname: Raveaux fullname: Raveaux, Romain – sequence: 6 givenname: Sébastien surname: Adam fullname: Adam, Sébastien |
| BackLink | https://doi.org/10.48550/arXiv.2410.01661$$DView paper in arXiv |
| BookMark | eNqFjr0OgkAQhK_Qwr8HsHJfQAQFYq8SCyu1v2yOBS6B5XKAytt7EHurSb58k5m5mHDNJMQ68L3wGEX-Du1Hv7x96IAfxHEwE49Ec6o5B4NtAcgpqF6VBKruuB14VtuqK5FAM-QWTdHAWzv1TGTgTpqdoKgibuFGaNl1lmKaYdnQ6pcLsUkuz9N1O85LY3WFtpfDDTneOPw3vh_-P2w |
| ContentType | Journal Article |
| Copyright | http://creativecommons.org/licenses/by/4.0 |
| Copyright_xml | – notice: http://creativecommons.org/licenses/by/4.0 |
| DBID | AKY GOX |
| DOI | 10.48550/arxiv.2410.01661 |
| DatabaseName | arXiv Computer Science arXiv.org |
| DatabaseTitleList | |
| Database_xml | – sequence: 1 dbid: GOX name: arXiv.org url: http://arxiv.org/find sourceTypes: Open Access Repository |
| DeliveryMethod | fulltext_linktorsrc |
| ExternalDocumentID | 2410_01661 |
| GroupedDBID | AKY GOX |
| ID | FETCH-arxiv_primary_2410_016613 |
| IEDL.DBID | GOX |
| IngestDate | Fri Oct 04 21:22:11 EDT 2024 |
| IsDoiOpenAccess | true |
| IsOpenAccess | true |
| IsPeerReviewed | false |
| IsScholarly | false |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-arxiv_primary_2410_016613 |
| OpenAccessLink | https://arxiv.org/abs/2410.01661 |
| ParticipantIDs | arxiv_primary_2410_01661 |
| PublicationCentury | 2000 |
| PublicationDate | 2024-10-02 |
| PublicationDateYYYYMMDD | 2024-10-02 |
| PublicationDate_xml | – month: 10 year: 2024 text: 2024-10-02 day: 02 |
| PublicationDecade | 2020 |
| PublicationYear | 2024 |
| Score | 3.874448 |
| SecondaryResourceType | preprint |
| Snippet | This paper presents Grammar Reinforcement Learning (GRL), a reinforcement
learning algorithm that uses Monte Carlo Tree Search (MCTS) and a transformer... |
| SourceID | arxiv |
| SourceType | Open Access Repository |
| SubjectTerms | Computer Science - Artificial Intelligence Computer Science - Formal Languages and Automata Theory |
| Title | Finding path and cycle counting formulae in graphs with Deep Reinforcement Learning |
| URI | https://arxiv.org/abs/2410.01661 |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| link | http://sdu.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwdV1NT8MwDLXYTlwQCND49oFrYW2SNT0itrITSIzDblWTOqgSmqaVIvj3JG4nuOxqR5GVRPKzYz8D3AplJk4LGWWVf8FSuzjSooqjkoTKhBe6NPQOzxfp81JPZ4EmB7e9MOXmu_7q-IFNc-_dy_jOg5IQ3wySJJRsPb0su89JpuLq1_-t8xiTRf-cRH4IBz26w4fuOo5gj1bHsMhrbh3BMP0XfeSO9sdrcTumAQNubD9KwnqFTCDdYEiP4pRoja_E3KaW03jY06G-n8BNPnt7nEdsRrHuOCOKYGHBFopTGPrInkaAxsSmJCsFGSdjqYwhlWllXZaObTXRZzDatcv5btUF7Cfe83LFWXIJw89NS1cwaKr2mo_vF6DNc5A |
| link.rule.ids | 228,230,782,887 |
| linkProvider | Cornell University |
| 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=Finding+path+and+cycle+counting+formulae+in+graphs+with+Deep+Reinforcement+Learning&rft.au=Piquenot%2C+Jason&rft.au=B%C3%A9rar%2C+Maxime&rft.au=H%C3%A9roux%2C+Pierre&rft.au=Ramel%2C+Jean-Yves&rft.date=2024-10-02&rft_id=info:doi/10.48550%2Farxiv.2410.01661&rft.externalDocID=2410_01661 |