SMAC-Hard: Enabling Mixed Opponent Strategy Script and Self-play on SMAC Article Swipe
Yue Deng
,
Yu Yan
,
Weiyu Ma
,
Zirui Wang
,
Wenhui Zhu
,
Jian Zhao
,
Y. Zhang
·
YOU?
·
· 2024
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2412.17707
YOU?
·
· 2024
· Open Access
·
· DOI: https://doi.org/10.48550/arxiv.2412.17707
The availability of challenging simulation environments is pivotal for advancing the field of Multi-Agent Reinforcement Learning (MARL). In cooperative MARL settings, the StarCraft Multi-Agent Challenge (SMAC) has gained prominence as a benchmark for algorithms following centralized training with decentralized execution paradigm. However, with continual advancements in SMAC, many algorithms now exhibit near-optimal performance, complicating the evaluation of their true effectiveness. To alleviate this problem, in this work, we highlight a critical issue: the default opponent policy in these environments lacks sufficient diversity, leading MARL algorithms to overfit and exploit unintended vulnerabilities rather than learning robust strategies. To overcome these limitations, we propose SMAC-HARD, a novel benchmark designed to enhance training robustness and evaluation comprehensiveness. SMAC-HARD supports customizable opponent strategies, randomization of adversarial policies, and interfaces for MARL self-play, enabling agents to generalize to varying opponent behaviors and improve model stability. Furthermore, we introduce a black-box testing framework wherein agents are trained without exposure to the edited opponent scripts but are tested against these scripts to evaluate the policy coverage and adaptability of MARL algorithms. We conduct extensive evaluations of widely used and state-of-the-art algorithms on SMAC-HARD, revealing the substantial challenges posed by edited and mixed strategy opponents. Additionally, the black-box strategy tests illustrate the difficulty of transferring learned policies to unseen adversaries. We envision SMAC-HARD as a critical step toward benchmarking the next generation of MARL algorithms, fostering progress in self-play methods for multi-agent systems. Our code is available at https://github.com/devindeng94/smac-hard.
Related Topics
Concepts
Metadata
- Type
- preprint
- Language
- en
- Landing Page
- http://arxiv.org/abs/2412.17707
- https://arxiv.org/pdf/2412.17707
- OA Status
- green
- Related Works
- 10
- OpenAlex ID
- https://openalex.org/W4405768320
All OpenAlex metadata
Raw OpenAlex JSON
- OpenAlex ID
-
https://openalex.org/W4405768320Canonical identifier for this work in OpenAlex
- DOI
-
https://doi.org/10.48550/arxiv.2412.17707Digital Object Identifier
- Title
-
SMAC-Hard: Enabling Mixed Opponent Strategy Script and Self-play on SMACWork title
- Type
-
preprintOpenAlex work type
- Language
-
enPrimary language
- Publication year
-
2024Year of publication
- Publication date
-
2024-12-23Full publication date if available
- Authors
-
Yue Deng, Yu Yan, Weiyu Ma, Zirui Wang, Wenhui Zhu, Jian Zhao, Y. ZhangList of authors in order
- Landing page
-
https://arxiv.org/abs/2412.17707Publisher landing page
- PDF URL
-
https://arxiv.org/pdf/2412.17707Direct 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/2412.17707Direct OA link when available
- Concepts
-
Computer science, Adversary, Computer securityTop concepts (fields/topics) attached by OpenAlex
- Cited by
-
0Total citation count in OpenAlex
- Related works (count)
-
10Other works algorithmically related by OpenAlex
Full payload
| id | https://openalex.org/W4405768320 |
|---|---|
| doi | https://doi.org/10.48550/arxiv.2412.17707 |
| ids.doi | https://doi.org/10.48550/arxiv.2412.17707 |
| ids.openalex | https://openalex.org/W4405768320 |
| fwci | |
| type | preprint |
| title | SMAC-Hard: Enabling Mixed Opponent Strategy Script and Self-play on SMAC |
| biblio.issue | |
| biblio.volume | |
| biblio.last_page | |
| biblio.first_page | |
| topics[0].id | https://openalex.org/T10320 |
| topics[0].field.id | https://openalex.org/fields/17 |
| topics[0].field.display_name | Computer Science |
| topics[0].score | 0.9545000195503235 |
| topics[0].domain.id | https://openalex.org/domains/3 |
| topics[0].domain.display_name | Physical Sciences |
| topics[0].subfield.id | https://openalex.org/subfields/1702 |
| topics[0].subfield.display_name | Artificial Intelligence |
| topics[0].display_name | Neural Networks and Applications |
| is_xpac | False |
| apc_list | |
| apc_paid | |
| concepts[0].id | https://openalex.org/C41008148 |
| concepts[0].level | 0 |
| concepts[0].score | 0.47906050086021423 |
| concepts[0].wikidata | https://www.wikidata.org/wiki/Q21198 |
| concepts[0].display_name | Computer science |
| concepts[1].id | https://openalex.org/C41065033 |
| concepts[1].level | 2 |
| concepts[1].score | 0.4353903830051422 |
| concepts[1].wikidata | https://www.wikidata.org/wiki/Q2825412 |
| concepts[1].display_name | Adversary |
| concepts[2].id | https://openalex.org/C38652104 |
| concepts[2].level | 1 |
| concepts[2].score | 0.18689879775047302 |
| concepts[2].wikidata | https://www.wikidata.org/wiki/Q3510521 |
| concepts[2].display_name | Computer security |
| keywords[0].id | https://openalex.org/keywords/computer-science |
| keywords[0].score | 0.47906050086021423 |
| keywords[0].display_name | Computer science |
| keywords[1].id | https://openalex.org/keywords/adversary |
| keywords[1].score | 0.4353903830051422 |
| keywords[1].display_name | Adversary |
| keywords[2].id | https://openalex.org/keywords/computer-security |
| keywords[2].score | 0.18689879775047302 |
| keywords[2].display_name | Computer security |
| language | en |
| locations[0].id | pmh:oai:arXiv.org:2412.17707 |
| 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 | |
| locations[0].pdf_url | https://arxiv.org/pdf/2412.17707 |
| locations[0].version | submittedVersion |
| locations[0].raw_type | text |
| locations[0].license_id | |
| locations[0].is_accepted | False |
| locations[0].is_published | False |
| locations[0].raw_source_name | |
| locations[0].landing_page_url | http://arxiv.org/abs/2412.17707 |
| locations[1].id | doi:10.48550/arxiv.2412.17707 |
| 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 | |
| locations[1].pdf_url | |
| locations[1].version | |
| locations[1].raw_type | article |
| locations[1].license_id | |
| 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.2412.17707 |
| indexed_in | arxiv, datacite |
| authorships[0].author.id | https://openalex.org/A5103063237 |
| authorships[0].author.orcid | https://orcid.org/0000-0002-4267-7664 |
| authorships[0].author.display_name | Yue Deng |
| authorships[0].author_position | first |
| authorships[0].raw_author_name | Deng, Yue |
| authorships[0].is_corresponding | False |
| authorships[1].author.id | https://openalex.org/A5100708610 |
| authorships[1].author.orcid | https://orcid.org/0000-0001-6697-5482 |
| authorships[1].author.display_name | Yu Yan |
| authorships[1].author_position | middle |
| authorships[1].raw_author_name | Yu, Yan |
| authorships[1].is_corresponding | False |
| authorships[2].author.id | https://openalex.org/A5058001011 |
| authorships[2].author.orcid | |
| authorships[2].author.display_name | Weiyu Ma |
| authorships[2].author_position | middle |
| authorships[2].raw_author_name | Ma, Weiyu |
| authorships[2].is_corresponding | False |
| authorships[3].author.id | https://openalex.org/A5100687847 |
| authorships[3].author.orcid | https://orcid.org/0000-0003-0626-742X |
| authorships[3].author.display_name | Zirui Wang |
| authorships[3].author_position | middle |
| authorships[3].raw_author_name | Wang, Zirui |
| authorships[3].is_corresponding | False |
| authorships[4].author.id | https://openalex.org/A5102752096 |
| authorships[4].author.orcid | |
| authorships[4].author.display_name | Wenhui Zhu |
| authorships[4].author_position | middle |
| authorships[4].raw_author_name | Zhu, Wenhui |
| authorships[4].is_corresponding | False |
| authorships[5].author.id | https://openalex.org/A5008612395 |
| authorships[5].author.orcid | https://orcid.org/0000-0002-9675-8267 |
| authorships[5].author.display_name | Jian Zhao |
| authorships[5].author_position | middle |
| authorships[5].raw_author_name | Zhao, Jian |
| authorships[5].is_corresponding | False |
| authorships[6].author.id | https://openalex.org/A5082169875 |
| authorships[6].author.orcid | |
| authorships[6].author.display_name | Y. Zhang |
| authorships[6].author_position | last |
| authorships[6].raw_author_name | Zhang, Yin |
| authorships[6].is_corresponding | False |
| has_content.pdf | False |
| has_content.grobid_xml | False |
| is_paratext | False |
| open_access.is_oa | True |
| open_access.oa_url | https://arxiv.org/pdf/2412.17707 |
| open_access.oa_status | green |
| open_access.any_repository_has_fulltext | False |
| created_date | 2025-10-10T00:00:00 |
| display_name | SMAC-Hard: Enabling Mixed Opponent Strategy Script and Self-play on SMAC |
| has_fulltext | False |
| is_retracted | False |
| updated_date | 2025-11-06T06:51:31.235846 |
| primary_topic.id | https://openalex.org/T10320 |
| primary_topic.field.id | https://openalex.org/fields/17 |
| primary_topic.field.display_name | Computer Science |
| primary_topic.score | 0.9545000195503235 |
| primary_topic.domain.id | https://openalex.org/domains/3 |
| primary_topic.domain.display_name | Physical Sciences |
| primary_topic.subfield.id | https://openalex.org/subfields/1702 |
| primary_topic.subfield.display_name | Artificial Intelligence |
| primary_topic.display_name | Neural Networks and Applications |
| related_works | https://openalex.org/W4391375266, https://openalex.org/W2899084033, https://openalex.org/W2748952813, https://openalex.org/W4388150944, https://openalex.org/W4242235492, https://openalex.org/W2390279801, https://openalex.org/W4237162029, https://openalex.org/W2367268135, https://openalex.org/W4391913857, https://openalex.org/W2358668433 |
| cited_by_count | 0 |
| locations_count | 2 |
| best_oa_location.id | pmh:oai:arXiv.org:2412.17707 |
| 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 | |
| best_oa_location.pdf_url | https://arxiv.org/pdf/2412.17707 |
| best_oa_location.version | submittedVersion |
| best_oa_location.raw_type | text |
| best_oa_location.license_id | |
| 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/2412.17707 |
| primary_location.id | pmh:oai:arXiv.org:2412.17707 |
| 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 | |
| primary_location.pdf_url | https://arxiv.org/pdf/2412.17707 |
| primary_location.version | submittedVersion |
| primary_location.raw_type | text |
| primary_location.license_id | |
| primary_location.is_accepted | False |
| primary_location.is_published | False |
| primary_location.raw_source_name | |
| primary_location.landing_page_url | http://arxiv.org/abs/2412.17707 |
| publication_date | 2024-12-23 |
| publication_year | 2024 |
| referenced_works_count | 0 |
| abstract_inverted_index.a | 30, 69, 103, 143, 216 |
| abstract_inverted_index.In | 17 |
| abstract_inverted_index.To | 60, 96 |
| abstract_inverted_index.We | 174, 212 |
| abstract_inverted_index.as | 29, 215 |
| abstract_inverted_index.at | 239 |
| abstract_inverted_index.by | 191 |
| abstract_inverted_index.in | 45, 64, 76, 229 |
| abstract_inverted_index.is | 6, 237 |
| abstract_inverted_index.of | 2, 12, 56, 120, 171, 178, 205, 224 |
| abstract_inverted_index.on | 184 |
| abstract_inverted_index.to | 85, 107, 130, 132, 153, 164, 209 |
| abstract_inverted_index.we | 67, 100, 141 |
| abstract_inverted_index.Our | 235 |
| abstract_inverted_index.The | 0 |
| abstract_inverted_index.and | 87, 111, 123, 136, 169, 181, 193 |
| abstract_inverted_index.are | 149, 159 |
| abstract_inverted_index.but | 158 |
| abstract_inverted_index.for | 8, 32, 125, 232 |
| abstract_inverted_index.has | 26 |
| abstract_inverted_index.now | 49 |
| abstract_inverted_index.the | 10, 21, 54, 72, 154, 166, 187, 198, 203, 221 |
| abstract_inverted_index.MARL | 19, 83, 126, 172, 225 |
| abstract_inverted_index.code | 236 |
| abstract_inverted_index.many | 47 |
| abstract_inverted_index.next | 222 |
| abstract_inverted_index.step | 218 |
| abstract_inverted_index.than | 92 |
| abstract_inverted_index.this | 62, 65 |
| abstract_inverted_index.true | 58 |
| abstract_inverted_index.used | 180 |
| abstract_inverted_index.with | 37, 42 |
| abstract_inverted_index.SMAC, | 46 |
| abstract_inverted_index.field | 11 |
| abstract_inverted_index.lacks | 79 |
| abstract_inverted_index.mixed | 194 |
| abstract_inverted_index.model | 138 |
| abstract_inverted_index.novel | 104 |
| abstract_inverted_index.posed | 190 |
| abstract_inverted_index.tests | 201 |
| abstract_inverted_index.their | 57 |
| abstract_inverted_index.these | 77, 98, 162 |
| abstract_inverted_index.work, | 66 |
| abstract_inverted_index.(SMAC) | 25 |
| abstract_inverted_index.agents | 129, 148 |
| abstract_inverted_index.edited | 155, 192 |
| abstract_inverted_index.gained | 27 |
| abstract_inverted_index.issue: | 71 |
| abstract_inverted_index.policy | 75, 167 |
| abstract_inverted_index.rather | 91 |
| abstract_inverted_index.robust | 94 |
| abstract_inverted_index.tested | 160 |
| abstract_inverted_index.toward | 219 |
| abstract_inverted_index.unseen | 210 |
| abstract_inverted_index.widely | 179 |
| abstract_inverted_index.(MARL). | 16 |
| abstract_inverted_index.against | 161 |
| abstract_inverted_index.conduct | 175 |
| abstract_inverted_index.default | 73 |
| abstract_inverted_index.enhance | 108 |
| abstract_inverted_index.exhibit | 50 |
| abstract_inverted_index.exploit | 88 |
| abstract_inverted_index.improve | 137 |
| abstract_inverted_index.leading | 82 |
| abstract_inverted_index.learned | 207 |
| abstract_inverted_index.methods | 231 |
| abstract_inverted_index.overfit | 86 |
| abstract_inverted_index.pivotal | 7 |
| abstract_inverted_index.propose | 101 |
| abstract_inverted_index.scripts | 157, 163 |
| abstract_inverted_index.testing | 145 |
| abstract_inverted_index.trained | 150 |
| abstract_inverted_index.varying | 133 |
| abstract_inverted_index.wherein | 147 |
| abstract_inverted_index.without | 151 |
| abstract_inverted_index.However, | 41 |
| abstract_inverted_index.Learning | 15 |
| abstract_inverted_index.coverage | 168 |
| abstract_inverted_index.critical | 70, 217 |
| abstract_inverted_index.designed | 106 |
| abstract_inverted_index.enabling | 128 |
| abstract_inverted_index.envision | 213 |
| abstract_inverted_index.evaluate | 165 |
| abstract_inverted_index.exposure | 152 |
| abstract_inverted_index.learning | 93 |
| abstract_inverted_index.opponent | 74, 117, 134, 156 |
| abstract_inverted_index.overcome | 97 |
| abstract_inverted_index.policies | 208 |
| abstract_inverted_index.problem, | 63 |
| abstract_inverted_index.progress | 228 |
| abstract_inverted_index.strategy | 195, 200 |
| abstract_inverted_index.supports | 115 |
| abstract_inverted_index.systems. | 234 |
| abstract_inverted_index.training | 36, 109 |
| abstract_inverted_index.Challenge | 24 |
| abstract_inverted_index.SMAC-HARD | 114, 214 |
| abstract_inverted_index.StarCraft | 22 |
| abstract_inverted_index.advancing | 9 |
| abstract_inverted_index.alleviate | 61 |
| abstract_inverted_index.available | 238 |
| abstract_inverted_index.behaviors | 135 |
| abstract_inverted_index.benchmark | 31, 105 |
| abstract_inverted_index.black-box | 144, 199 |
| abstract_inverted_index.continual | 43 |
| abstract_inverted_index.execution | 39 |
| abstract_inverted_index.extensive | 176 |
| abstract_inverted_index.following | 34 |
| abstract_inverted_index.fostering | 227 |
| abstract_inverted_index.framework | 146 |
| abstract_inverted_index.highlight | 68 |
| abstract_inverted_index.introduce | 142 |
| abstract_inverted_index.paradigm. | 40 |
| abstract_inverted_index.policies, | 122 |
| abstract_inverted_index.revealing | 186 |
| abstract_inverted_index.self-play | 230 |
| abstract_inverted_index.settings, | 20 |
| abstract_inverted_index.SMAC-HARD, | 102, 185 |
| abstract_inverted_index.algorithms | 33, 48, 84, 183 |
| abstract_inverted_index.challenges | 189 |
| abstract_inverted_index.difficulty | 204 |
| abstract_inverted_index.diversity, | 81 |
| abstract_inverted_index.evaluation | 55, 112 |
| abstract_inverted_index.generalize | 131 |
| abstract_inverted_index.generation | 223 |
| abstract_inverted_index.illustrate | 202 |
| abstract_inverted_index.interfaces | 124 |
| abstract_inverted_index.opponents. | 196 |
| abstract_inverted_index.prominence | 28 |
| abstract_inverted_index.robustness | 110 |
| abstract_inverted_index.self-play, | 127 |
| abstract_inverted_index.simulation | 4 |
| abstract_inverted_index.stability. | 139 |
| abstract_inverted_index.sufficient | 80 |
| abstract_inverted_index.unintended | 89 |
| abstract_inverted_index.Multi-Agent | 13, 23 |
| abstract_inverted_index.adversarial | 121 |
| abstract_inverted_index.algorithms, | 226 |
| abstract_inverted_index.algorithms. | 173 |
| abstract_inverted_index.centralized | 35 |
| abstract_inverted_index.challenging | 3 |
| abstract_inverted_index.cooperative | 18 |
| abstract_inverted_index.evaluations | 177 |
| abstract_inverted_index.multi-agent | 233 |
| abstract_inverted_index.strategies, | 118 |
| abstract_inverted_index.strategies. | 95 |
| abstract_inverted_index.substantial | 188 |
| abstract_inverted_index.Furthermore, | 140 |
| abstract_inverted_index.adaptability | 170 |
| abstract_inverted_index.advancements | 44 |
| abstract_inverted_index.adversaries. | 211 |
| abstract_inverted_index.availability | 1 |
| abstract_inverted_index.benchmarking | 220 |
| abstract_inverted_index.complicating | 53 |
| abstract_inverted_index.customizable | 116 |
| abstract_inverted_index.environments | 5, 78 |
| abstract_inverted_index.limitations, | 99 |
| abstract_inverted_index.near-optimal | 51 |
| abstract_inverted_index.performance, | 52 |
| abstract_inverted_index.transferring | 206 |
| abstract_inverted_index.Additionally, | 197 |
| abstract_inverted_index.Reinforcement | 14 |
| abstract_inverted_index.decentralized | 38 |
| abstract_inverted_index.randomization | 119 |
| abstract_inverted_index.effectiveness. | 59 |
| abstract_inverted_index.vulnerabilities | 90 |
| abstract_inverted_index.state-of-the-art | 182 |
| abstract_inverted_index.comprehensiveness. | 113 |
| abstract_inverted_index.https://github.com/devindeng94/smac-hard. | 240 |
| cited_by_percentile_year | |
| countries_distinct_count | 0 |
| institutions_distinct_count | 7 |
| citation_normalized_percentile |