MACS: Multi-Agent Reinforcement Learning for Optimization of Crystal Structures Article Swipe
YOU?
·
· 2025
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2506.04195
Geometry optimization of atomic structures is a common and crucial task in computational chemistry and materials design. Following the learning to optimize paradigm, we propose a new multi-agent reinforcement learning method called Multi-Agent Crystal Structure optimization (MACS) to address periodic crystal structure optimization. MACS treats geometry optimization as a partially observable Markov game in which atoms are agents that adjust their positions to collectively discover a stable configuration. We train MACS across various compositions of reported crystalline materials to obtain a policy that successfully optimizes structures from the training compositions as well as structures of larger sizes and unseen compositions, confirming its excellent scalability and zero-shot transferability. We benchmark our approach against a broad range of state-of-the-art optimization methods and demonstrate that MACS optimizes periodic crystal structures significantly faster, with fewer energy calculations, and the lowest failure rate.
Related Topics
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2506.04195
- https://arxiv.org/pdf/2506.04195
- OA Status
- green
- OpenAlex ID
- https://openalex.org/W4416075339
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4416075339Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2506.04195Digital Object Identifier
- Title
-
MACS: Multi-Agent Reinforcement Learning for Optimization of Crystal StructuresWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2025Year of publication
- Publication date
-
2025-06-04Full publication date if available
- Authors
-
Elena Zamaraeva, Christopher M. Collins, George R. Darling, Matthew S. Dyer, Bei Peng, Rahul Savani, Dmytro Antypov, Vladimir V. Gusev, Judith Clymo, Paul G. Spirakis, Matthew J. RosseinskyList of authors in order
- Landing page
-
https://arxiv.org/abs/2506.04195Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2506.04195Direct link to full text PDF
- Open access
-
YesWhether a free full text is available
- OA status
-
greenOpen access status per OpenAlex
- OA URL
-
https://arxiv.org/pdf/2506.04195Direct OA link when available
- Cited by
-
0Total citation count in OpenAlex
Full payload
| id | https://openalex.org/W4416075339 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2506.04195 |
| ids.doi | https://doi.org/10.48550/arxiv.2506.04195 |
| ids.openalex | https://openalex.org/W4416075339 |
| fwci | |
| type | preprint |
| title | MACS: Multi-Agent Reinforcement Learning for Optimization of Crystal Structures |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2506.04195 |
| locations[0].is_oa | True |
| locations[0].source.id | https://openalex.org/S4306400194 |
| locations[0].source.issn | |
| locations[0].source.type | repository |
| locations[0].source.is_oa | True |
| locations[0].source.issn_l | |
| locations[0].source.is_core | False |
| locations[0].source.is_in_doaj | False |
| locations[0].source.display_name | arXiv (Cornell University) |
| locations[0].source.host_organization | https://openalex.org/I205783295 |
| locations[0].source.host_organization_name | Cornell University |
| locations[0].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[0].license | cc-by |
| locations[0].pdf_url | https://arxiv.org/pdf/2506.04195 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | https://openalex.org/licenses/cc-by |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2506.04195 |
| locations[1].id | doi:10.48550/arxiv.2506.04195 |
| locations[1].is_oa | True |
| locations[1].source.id | https://openalex.org/S4306400194 |
| locations[1].source.issn | |
| locations[1].source.type | repository |
| locations[1].source.is_oa | True |
| locations[1].source.issn_l | |
| locations[1].source.is_core | False |
| locations[1].source.is_in_doaj | False |
| locations[1].source.display_name | arXiv (Cornell University) |
| locations[1].source.host_organization | https://openalex.org/I205783295 |
| locations[1].source.host_organization_name | Cornell University |
| locations[1].source.host_organization_lineage | https://openalex.org/I205783295 |
| locations[1].license | cc-by |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | https://openalex.org/licenses/cc-by |
| locations[1].is_accepted | False |
| locations[1].is_published | |
| locations[1].raw_source_name | |
| locations[1].landing_page_url | https://doi.org/10.48550/arxiv.2506.04195 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5062930386 |
| authorships[0].author.orcid | https://orcid.org/0000-0002-7948-2641 |
| authorships[0].author.display_name | Elena Zamaraeva |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Zamaraeva, Elena |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5036142142 |
| authorships[1].author.orcid | https://orcid.org/0000-0002-0101-4426 |
| authorships[1].author.display_name | Christopher M. Collins |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Collins, Christopher M. |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5080860663 |
| authorships[2].author.orcid | https://orcid.org/0000-0001-9329-9993 |
| authorships[2].author.display_name | George R. Darling |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Darling, George R. |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5091597124 |
| authorships[3].author.orcid | https://orcid.org/0000-0002-4923-3003 |
| authorships[3].author.display_name | Matthew S. Dyer |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Dyer, Matthew S. |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5101481833 |
| authorships[4].author.orcid | https://orcid.org/0000-0003-0152-3180 |
| authorships[4].author.display_name | Bei Peng |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Peng, Bei |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5084692264 |
| authorships[5].author.orcid | https://orcid.org/0000-0003-1262-7831 |
| authorships[5].author.display_name | Rahul Savani |
| authorships[5].author_position | middle |
| authorships[5].raw_author_name | Savani, Rahul |
| authorships[5].is_corresponding | False |
| authorships[6].author.id | https://openalex.org/A5062223660 |
| authorships[6].author.orcid | https://orcid.org/0000-0003-1893-7785 |
| authorships[6].author.display_name | Dmytro Antypov |
| authorships[6].author_position | middle |
| authorships[6].raw_author_name | Antypov, Dmytro |
| authorships[6].is_corresponding | False |
| authorships[7].author.id | https://openalex.org/A5067994724 |
| authorships[7].author.orcid | https://orcid.org/0000-0002-2815-607X |
| authorships[7].author.display_name | Vladimir V. Gusev |
| authorships[7].author_position | middle |
| authorships[7].raw_author_name | Gusev, Vladimir V. |
| authorships[7].is_corresponding | False |
| authorships[8].author.id | https://openalex.org/A5114353954 |
| authorships[8].author.orcid | |
| authorships[8].author.display_name | Judith Clymo |
| authorships[8].author_position | middle |
| authorships[8].raw_author_name | Clymo, Judith |
| authorships[8].is_corresponding | False |
| authorships[9].author.id | https://openalex.org/A5011756177 |
| authorships[9].author.orcid | https://orcid.org/0000-0001-5396-3749 |
| authorships[9].author.display_name | Paul G. Spirakis |
| authorships[9].author_position | middle |
| authorships[9].raw_author_name | Spirakis, Paul G. |
| authorships[9].is_corresponding | False |
| authorships[10].author.id | https://openalex.org/A5054755054 |
| authorships[10].author.orcid | https://orcid.org/0000-0002-1910-2483 |
| authorships[10].author.display_name | Matthew J. Rosseinsky |
| authorships[10].author_position | last |
| authorships[10].raw_author_name | Rosseinsky, Matthew J. |
| authorships[10].is_corresponding | False |
| has_content.pdf | True |
| has_content.grobid_xml | True |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2506.04195 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | MACS: Multi-Agent Reinforcement Learning for Optimization of Crystal Structures |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-28T09:52:02.233439 |
| primary_topic | |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2506.04195 |
| best_oa_location.is_oa | True |
| best_oa_location.source.id | https://openalex.org/S4306400194 |
| best_oa_location.source.issn | |
| best_oa_location.source.type | repository |
| best_oa_location.source.is_oa | True |
| best_oa_location.source.issn_l | |
| best_oa_location.source.is_core | False |
| best_oa_location.source.is_in_doaj | False |
| best_oa_location.source.display_name | arXiv (Cornell University) |
| best_oa_location.source.host_organization | https://openalex.org/I205783295 |
| best_oa_location.source.host_organization_name | Cornell University |
| best_oa_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| best_oa_location.license | cc-by |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2506.04195 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | https://openalex.org/licenses/cc-by |
| best_oa_location.is_accepted | False |
| best_oa_location.is_published | False |
| best_oa_location.raw_source_name | |
| best_oa_location.landing_page_url | http://arxiv.org/abs/2506.04195 |
| primary_location.id | pmh:oai:arXiv.org:2506.04195 |
| primary_location.is_oa | True |
| primary_location.source.id | https://openalex.org/S4306400194 |
| primary_location.source.issn | |
| primary_location.source.type | repository |
| primary_location.source.is_oa | True |
| primary_location.source.issn_l | |
| primary_location.source.is_core | False |
| primary_location.source.is_in_doaj | False |
| primary_location.source.display_name | arXiv (Cornell University) |
| primary_location.source.host_organization | https://openalex.org/I205783295 |
| primary_location.source.host_organization_name | Cornell University |
| primary_location.source.host_organization_lineage | https://openalex.org/I205783295 |
| primary_location.license | cc-by |
| primary_location.pdf_url | https://arxiv.org/pdf/2506.04195 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | https://openalex.org/licenses/cc-by |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2506.04195 |
| publication_date | 2025-06-04 |
| publication_year | 2025 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 6, 25, 48, 65, 80, 112 |
| abstract_inverted_index.We | 68, 107 |
| abstract_inverted_index.as | 47, 90, 92 |
| abstract_inverted_index.in | 11, 53 |
| abstract_inverted_index.is | 5 |
| abstract_inverted_index.of | 2, 74, 94, 115 |
| abstract_inverted_index.to | 20, 37, 62, 78 |
| abstract_inverted_index.we | 23 |
| abstract_inverted_index.and | 8, 14, 97, 104, 119, 133 |
| abstract_inverted_index.are | 56 |
| abstract_inverted_index.its | 101 |
| abstract_inverted_index.new | 26 |
| abstract_inverted_index.our | 109 |
| abstract_inverted_index.the | 18, 87, 134 |
| abstract_inverted_index.MACS | 43, 70, 122 |
| abstract_inverted_index.from | 86 |
| abstract_inverted_index.game | 52 |
| abstract_inverted_index.task | 10 |
| abstract_inverted_index.that | 58, 82, 121 |
| abstract_inverted_index.well | 91 |
| abstract_inverted_index.with | 129 |
| abstract_inverted_index.atoms | 55 |
| abstract_inverted_index.broad | 113 |
| abstract_inverted_index.fewer | 130 |
| abstract_inverted_index.range | 114 |
| abstract_inverted_index.rate. | 137 |
| abstract_inverted_index.sizes | 96 |
| abstract_inverted_index.their | 60 |
| abstract_inverted_index.train | 69 |
| abstract_inverted_index.which | 54 |
| abstract_inverted_index.(MACS) | 36 |
| abstract_inverted_index.Markov | 51 |
| abstract_inverted_index.across | 71 |
| abstract_inverted_index.adjust | 59 |
| abstract_inverted_index.agents | 57 |
| abstract_inverted_index.atomic | 3 |
| abstract_inverted_index.called | 31 |
| abstract_inverted_index.common | 7 |
| abstract_inverted_index.energy | 131 |
| abstract_inverted_index.larger | 95 |
| abstract_inverted_index.lowest | 135 |
| abstract_inverted_index.method | 30 |
| abstract_inverted_index.obtain | 79 |
| abstract_inverted_index.policy | 81 |
| abstract_inverted_index.stable | 66 |
| abstract_inverted_index.treats | 44 |
| abstract_inverted_index.unseen | 98 |
| abstract_inverted_index.Crystal | 33 |
| abstract_inverted_index.address | 38 |
| abstract_inverted_index.against | 111 |
| abstract_inverted_index.crucial | 9 |
| abstract_inverted_index.crystal | 40, 125 |
| abstract_inverted_index.design. | 16 |
| abstract_inverted_index.failure | 136 |
| abstract_inverted_index.faster, | 128 |
| abstract_inverted_index.methods | 118 |
| abstract_inverted_index.propose | 24 |
| abstract_inverted_index.various | 72 |
| abstract_inverted_index.Geometry | 0 |
| abstract_inverted_index.approach | 110 |
| abstract_inverted_index.discover | 64 |
| abstract_inverted_index.geometry | 45 |
| abstract_inverted_index.learning | 19, 29 |
| abstract_inverted_index.optimize | 21 |
| abstract_inverted_index.periodic | 39, 124 |
| abstract_inverted_index.reported | 75 |
| abstract_inverted_index.training | 88 |
| abstract_inverted_index.Following | 17 |
| abstract_inverted_index.Structure | 34 |
| abstract_inverted_index.benchmark | 108 |
| abstract_inverted_index.chemistry | 13 |
| abstract_inverted_index.excellent | 102 |
| abstract_inverted_index.materials | 15, 77 |
| abstract_inverted_index.optimizes | 84, 123 |
| abstract_inverted_index.paradigm, | 22 |
| abstract_inverted_index.partially | 49 |
| abstract_inverted_index.positions | 61 |
| abstract_inverted_index.structure | 41 |
| abstract_inverted_index.zero-shot | 105 |
| abstract_inverted_index.confirming | 100 |
| abstract_inverted_index.observable | 50 |
| abstract_inverted_index.structures | 4, 85, 93, 126 |
| abstract_inverted_index.Multi-Agent | 32 |
| abstract_inverted_index.crystalline | 76 |
| abstract_inverted_index.demonstrate | 120 |
| abstract_inverted_index.multi-agent | 27 |
| abstract_inverted_index.scalability | 103 |
| abstract_inverted_index.collectively | 63 |
| abstract_inverted_index.compositions | 73, 89 |
| abstract_inverted_index.optimization | 1, 35, 46, 117 |
| abstract_inverted_index.successfully | 83 |
| abstract_inverted_index.calculations, | 132 |
| abstract_inverted_index.compositions, | 99 |
| abstract_inverted_index.computational | 12 |
| abstract_inverted_index.optimization. | 42 |
| abstract_inverted_index.reinforcement | 28 |
| abstract_inverted_index.significantly | 127 |
| abstract_inverted_index.configuration. | 67 |
| abstract_inverted_index.state-of-the-art | 116 |
| abstract_inverted_index.transferability. | 106 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 11 |
| citation_normalized_percentile |