{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/uncovering-coresets-for-classification-with","title":"Uncovering Coresets for Classification With Multi-Objective Evolutionary Algorithms","arxiv_id":"2002.08645","date":"2020-02-20","proceeding":null,"authors":["Pietro Barbiero","Giovanni Squillero","Alberto Tonda"],"abstract":"A coreset is a subset of the training set, using which a machine learning algorithm obtains performances similar to what it would deliver if trained over the whole original data. Coreset discovery is an active and open line of research as it allows improving training speed for the algorithms and may help human understanding the results. Building on previous works, a novel approach is presented: candidate corsets are iteratively optimized, adding and removing samples. As there is an obvious trade-off between limiting training size and quality of the results, a multi-objective evolutionary algorithm is used to minimize simultaneously the number of points in the set and the classification error. Experimental results on non-trivial benchmarks show that the proposed approach is able to deliver results that allow a classifier to obtain lower error and better ability of generalizing on unseen data than state-of-the-art coreset discovery techniques.","url_abs":"https://arxiv.org/abs/2002.08645v1","url_pdf":"https://arxiv.org/pdf/2002.08645v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"uncovering-coresets-for-classification-with","repo_url":"https://github.com/pietrobarbiero/meco","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"core-set-discovery","task_name":"Core set discovery"},{"task_slug":"evolutionary-algorithms","task_name":"Evolutionary Algorithms"},{"task_slug":"classification","task_name":"General Classification"}],"methods":[{"method_slug":"speed","method_name":"SPEED"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/core-set-discovery-on-abalone","task":"Core set discovery","dataset":"Abalone","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"18.6"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-amazon-employee-access","task":"Core set discovery","dataset":"Amazon-employee-access","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"91.5"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-credit-g","task":"Core set discovery","dataset":"Credit-g","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"74.3"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-electricity","task":"Core set discovery","dataset":"Electricity","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"69.3"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-glass-identification","task":"Core set discovery","dataset":"Glass identification","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"64.3"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-isolet","task":"Core set discovery","dataset":"ISOLET","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"90.5"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-jm1","task":"Core set discovery","dataset":"JM1","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"77.1"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-kr-vs-kp","task":"Core set discovery","dataset":"Kr-vs-kp","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"93.7"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-letter","task":"Core set discovery","dataset":"Letter","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"65.9"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-mnist","task":"Core set discovery","dataset":"MNIST","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"77.2"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-mozilla4","task":"Core set discovery","dataset":"Mozilla4","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"91.2"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-soybean","task":"Core set discovery","dataset":"Soybean","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"91.1"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-uci-gas","task":"Core set discovery","dataset":"UCI GAS","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"94.6"},"uses_additional_data":false},{"leaderboard":"/sota/core-set-discovery-on-micro-mass","task":"Core set discovery","dataset":"micro-mass","model":"EvoCore","rank_in_archive_order":1,"of":1,"metrics":{"F1(10-fold)":"83.9"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2002.08645","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}