{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T17:48:11Z","timestamp":1782841691990,"version":"3.54.5"},"reference-count":47,"publisher":"Association for Computing Machinery (ACM)","issue":"2","license":[{"start":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T00:00:00Z","timestamp":1782777600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Evol. Learn. Optim."],"published-print":{"date-parts":[[2026,6,30]]},"abstract":"<jats:p>\n                    In complex control tasks, finding high-performing policies often requires discovering and exploiting specific behavioral strategies. While Quality-Diversity (QD) algorithms can uncover these strategies through extensive behavior space exploration, they sacrifice efficiency by improving suboptimal behaviors. Conversely, Evolution Strategies (ES) achieve impressive performance through focused optimization but frequently become trapped in local optima due to limited behavioral exploration. We present Quality with\n                    <jats:italic toggle=\"yes\">J<\/jats:italic>\n                    ust\n                    <jats:italic toggle=\"yes\">E<\/jats:italic>\n                    nough\n                    <jats:italic toggle=\"yes\">Di<\/jats:italic>\n                    versity (JEDi), a new optimization framework that resolves this fundamental tension. JEDi employs a combination of Gaussian Process modeling and parallel ES to intelligently explore behavioral space while maintaining focused optimization. At its core, JEDi learns a probabilistic mapping between behaviors and performance, using this model to identify and target promising behavioral regions that could unlock better solutions. This targeted exploration is achieved through multiple Evolution Strategy emitters that simultaneously optimize toward selected behaviors while maximizing task performance. To further improve JEDi\u2019s exploration capabilities, we introduce its\n                    <jats:italic toggle=\"yes\">Dynamic<\/jats:italic>\n                    variant DyJEDi with an adaptive restart mechanism that dynamically detects and responds to emitter convergence, independently restarting each Evolution Strategy when it stagnates in both behavior and fitness space. This dynamic approach significantly improves exploration efficiency and robustness to local optima. We demonstrate that DyJEDi outperforms both traditional ES and QD approaches across challenging continuous robotics control tasks, achieving higher final performance. Most notably, DyJEDi solves several hard exploration problems where standard ES methods consistently fail.\n                  <\/jats:p>","DOI":"10.1145\/3818608","type":"journal-article","created":{"date-parts":[[2026,5,28]],"date-time":"2026-05-28T13:31:37Z","timestamp":1779975097000},"page":"1-24","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Restart of the JEDi: Dynamic Quality with Just Enough Diversity"],"prefix":"10.1145","volume":"6","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8919-8639","authenticated-orcid":false,"given":"Paul","family":"Templier","sequence":"first","affiliation":[{"name":"Imperial College London, London, United Kingdom of Great Britain and Northern Ireland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4539-8211","authenticated-orcid":false,"given":"Luca","family":"Grillotti","sequence":"additional","affiliation":[{"name":"Imperial College London, London, United Kingdom of Great Britain and Northern Ireland"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8559-1617","authenticated-orcid":false,"given":"Emmanuel","family":"Rachelson","sequence":"additional","affiliation":[{"name":"F\u00e9d\u00e9ration ENAC ISAE-SUPAERO ONERA, Universit\u00e9 de Toulouse, Toulouse, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2414-0051","authenticated-orcid":false,"given":"Dennis G.","family":"Wilson","sequence":"additional","affiliation":[{"name":"F\u00e9d\u00e9ration ENAC ISAE-SUPAERO ONERA, Universit\u00e9 de Toulouse, Toulouse, France"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3190-7073","authenticated-orcid":false,"given":"Antoine","family":"Cully","sequence":"additional","affiliation":[{"name":"Imperial College London, London, United Kingdom of Great Britain and Northern Ireland"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,6,30]]},"reference":[{"key":"e_1_3_2_2_1","doi-asserted-by":"publisher","DOI":"10.1109\/CEC.2005.1554902"},{"key":"e_1_3_2_3_1","doi-asserted-by":"publisher","DOI":"10.1145\/1830483.1830504"},{"key":"e_1_3_2_4_1","volume-title":"Proceedings of the 39th International Conference on Machine Learning","author":"Calandriello Daniele","year":"2022","unstructured":"Daniele Calandriello, Luigi Carratino, Alessandro Lazaric, Michal Valko, and Lorenzo Rosasco. 2022. Scaling gaussian process optimization by evaluating a few unique candidates multiple times. In Proceedings of the 39th International Conference on Machine Learning."},{"key":"e_1_3_2_5_1","doi-asserted-by":"crossref","unstructured":"Patryk Chrabaszcz Ilya Loshchilov and Frank Hutter. 2018. Back to Basics: Benchmarking Canonical Evolution Strategies for Playing Atari. arXiv:1802.08842. Retrieved from https:\/\/arxiv.org\/abs\/1802.08842","DOI":"10.24963\/ijcai.2018\/197"},{"key":"e_1_3_2_6_1","doi-asserted-by":"publisher","DOI":"10.1145\/3377930.3390217"},{"key":"e_1_3_2_7_1","unstructured":"Edoardo Conti Vashisht Madhavan Felipe Petroski Such Joel Lehman Kenneth O. Stanley and Jeff Clune. 2018. Improving exploration in evolution strategies for deep reinforcement learning via a population of novelty-seeking agents. arXiv:1712.06560. Retrieved from https:\/\/arxiv.org\/abs\/1712.06560"},{"key":"e_1_3_2_8_1","doi-asserted-by":"publisher","DOI":"10.1145\/3449639.3459326"},{"key":"e_1_3_2_9_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature14422"},{"key":"e_1_3_2_10_1","doi-asserted-by":"publisher","DOI":"10.5555\/2627435.2750368"},{"key":"e_1_3_2_11_1","unstructured":"Maxence Faldor F\u00e9lix Chalumeau Manon Flageat and Antoine Cully. 2023. Synergizing quality-diversity with descriptor-conditioned reinforcement learning. arXiv:2401.08632. Retrieved from https:\/\/arxiv.org\/abs\/2401.08632"},{"key":"e_1_3_2_12_1","doi-asserted-by":"publisher","unstructured":"Maxence Faldor Robert Tjarko Lange and Antoine Cully. 2025. Discovering quality-diversity algorithms via meta-black-box optimization. DOI: 10.48550\/arXiv.2502.02190","DOI":"10.48550\/arXiv.2502.02190"},{"key":"e_1_3_2_13_1","volume-title":"Real-Parameter Black-Box Optimization Benchmarking 2009: Presentation of the Noiseless Functions","author":"Finck Steffen","year":"2010","unstructured":"Steffen Finck, Nikolaus Hansen, Raymond Ros, and Anne Auger. 2010. Real-Parameter Black-Box Optimization Benchmarking 2009: Presentation of the Noiseless Functions. Technical Report. Citeseer."},{"key":"e_1_3_2_14_1","doi-asserted-by":"publisher","DOI":"10.1162\/isal_a_00316"},{"key":"e_1_3_2_15_1","unstructured":"Manon Flageat and Antoine Cully. 2023. Uncertain quality-diversity: Evaluation methodology and new methods for quality-diversity in uncertain domains. arXiv:2302.00463. Retrieved from https:\/\/arxiv.org\/abs\/2302.00463"},{"key":"e_1_3_2_16_1","doi-asserted-by":"crossref","unstructured":"Manon Flageat Bryan Lim and Antoine Cully. 2023. Multiple hands make light work: Enhancing quality and diversity using MAP-Elites with multiple parallel evolution strategies. arXiv:2303.06137. Retrieved from https:\/\/arxiv.org\/abs\/2303.06137","DOI":"10.1145\/3638529.3654089"},{"key":"e_1_3_2_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/3638529.3654089"},{"key":"e_1_3_2_18_1","doi-asserted-by":"crossref","unstructured":"Matthew C. Fontaine and Stefanos Nikolaidis. 2023. Covariance matrix adaptation MAP-annealing. arXiv:2205.10752. Retrieved from https:\/\/arxiv.org\/abs\/2205.10752","DOI":"10.1145\/3583131.3590389"},{"key":"e_1_3_2_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/3377930.3390232"},{"key":"e_1_3_2_20_1","unstructured":"C. Daniel Freeman Erik Frey Anton Raichuk Sertan Girgin Igor Mordatch and Olivier Bachem. 2021. Brax\u2014A Differentiable Physics Engine for Large Scale Rigid Body Simulation. arXiv:2106.13281. Retrieved from https:\/\/arxiv.org\/abs\/2106.13281"},{"key":"e_1_3_2_21_1","doi-asserted-by":"publisher","DOI":"10.1162\/evco_a_00231"},{"key":"e_1_3_2_22_1","doi-asserted-by":"publisher","DOI":"10.1145\/3583133.3596387"},{"key":"e_1_3_2_23_1","doi-asserted-by":"publisher","DOI":"10.1145\/3583131.3590498"},{"key":"e_1_3_2_24_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58112-1_10"},{"key":"e_1_3_2_25_1","unstructured":"Nikolaus Hansen. 2016. The CMA evolution strategy: A tutorial. arXiv:1604.00772. Retrieved from https:\/\/arxiv.org\/abs\/1604.00772"},{"key":"e_1_3_2_26_1","doi-asserted-by":"publisher","DOI":"10.1145\/3583131.3590486"},{"key":"e_1_3_2_27_1","unstructured":"Paul Kent Adam Gaier Jean-Baptiste Mouret and Juergen Branke. 2023. BOP-Elites a Bayesian optimisation approach to quality diversity search with black-box descriptor functions. arXiv:2307.09326. Retrieved from https:\/\/arxiv.org\/abs\/2307.09326"},{"key":"e_1_3_2_28_1","unstructured":"Robert Tjarko Lange. 2022. Evosax: JAX-based Evolution Strategies. arXiv:2212.04180. Retrieved from https:\/\/arxiv.org\/abs\/2212.04180"},{"key":"e_1_3_2_29_1","doi-asserted-by":"publisher","unstructured":"Joel Lehman and Kenneth O. Stanley. 2011. Abandoning Objectives: Evolution through the Search for Novelty Alone. 39. DOI: 10.1162\/EVCO_a_00025","DOI":"10.1162\/EVCO_a_00025"},{"key":"e_1_3_2_30_1","unstructured":"Bryan Lim Maxime Allard Luca Grillotti and Antoine Cully. 2022. Accelerated quality-diversity for robotics through massive parallelism. arXiv:2202.01258. Retrieved from https:\/\/arxiv.org\/abs\/2202.01258"},{"key":"e_1_3_2_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/CEC.2013.6557593"},{"key":"e_1_3_2_32_1","unstructured":"Ilya Loshchilov Tobias Glasmachers and Hans-Georg Beyer. 2017. Limited-memory matrix adaptation for large scale black-box optimization. arXiv:1705.06693. Retrieved from https:\/\/arxiv.org\/abs\/1705.06693"},{"key":"e_1_3_2_33_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-32937-1_30"},{"key":"e_1_3_2_34_1","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177730491"},{"key":"e_1_3_2_35_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-18272-3_10"},{"key":"e_1_3_2_36_1","unstructured":"Jean-Baptiste Mouret and Jeff Clune. 2015. Illuminating search spaces by mapping Elites. arXiv:1504.04909. Retrieved from https:\/\/arxiv.org\/abs\/1504.04909"},{"key":"e_1_3_2_37_1","doi-asserted-by":"publisher","DOI":"10.1162\/EVCO_a_00048"},{"key":"e_1_3_2_38_1","doi-asserted-by":"publisher","DOI":"10.1145\/3449639.3459304"},{"key":"e_1_3_2_39_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-81283-5_8"},{"key":"e_1_3_2_40_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-87700-4_30"},{"key":"e_1_3_2_41_1","unstructured":"Tim Salimans Jonathan Ho Xi Chen Szymon Sidor and Ilya Sutskever. 2017. Evolution Strategies as a Scalable Alternative to Reinforcement Learning. arXiv:1703.03864. Retrieved from https:\/\/arxiv.org\/abs\/1703.03864"},{"key":"e_1_3_2_42_1","doi-asserted-by":"publisher","DOI":"10.1145\/3449639.3459321"},{"key":"e_1_3_2_43_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2011.2182033"},{"key":"e_1_3_2_44_1","doi-asserted-by":"publisher","DOI":"10.1145\/3638529.3654047"},{"key":"e_1_3_2_45_1","unstructured":"Bryon Tjanaka Matthew C. Fontaine David H. Lee Aniruddha Kalkar and Stefanos Nikolaidis. 2023. Training diverse high-dimensional controllers by scaling covariance matrix adaptation MAP-annealing. arXiv:2210.02622. Retrieved from https:\/\/arxiv.org\/abs\/2210.02622"},{"key":"e_1_3_2_46_1","doi-asserted-by":"crossref","unstructured":"Bryon Tjanaka Matthew C. Fontaine Julian Togelius and Stefanos Nikolaidis. 2022. Approximating gradients for differentiable quality diversity in reinforcement learning. arXiv:2202.03666. Retrieved from https:\/\/arxiv.org\/abs\/2202.03666","DOI":"10.1145\/3512290.3528705"},{"key":"e_1_3_2_47_1","doi-asserted-by":"crossref","unstructured":"Yuxing Wang Tiantian Zhang Yongzhe Chang Bin Liang Xueqian Wang and Bo Yuan. 2022. A surrogate-assisted controller for expensive evolutionary reinforcement learning. arXiv:2201.00129. Retrieved from https:\/\/arxiv.org\/abs\/2201.00129","DOI":"10.1016\/j.ins.2022.10.134"},{"key":"e_1_3_2_48_1","doi-asserted-by":"publisher","DOI":"10.1145\/3512290.3528718"}],"container-title":["ACM Transactions on Evolutionary Learning and Optimization"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3818608","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T17:16:26Z","timestamp":1782839786000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3818608"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,30]]},"references-count":47,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2026,6,30]]}},"alternative-id":["10.1145\/3818608"],"URL":"https:\/\/doi.org\/10.1145\/3818608","relation":{},"ISSN":["2688-299X","2688-3007"],"issn-type":[{"value":"2688-299X","type":"print"},{"value":"2688-3007","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,30]]},"assertion":[{"value":"2025-01-24","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-04-25","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-06-30","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}