{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,22]],"date-time":"2026-04-22T19:12:21Z","timestamp":1776885141659,"version":"3.51.2"},"reference-count":26,"publisher":"Association for Computing Machinery (ACM)","issue":"8","license":[{"start":{"date-parts":[[2021,7,26]],"date-time":"2021-07-26T00:00:00Z","timestamp":1627257600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["Commun. ACM"],"published-print":{"date-parts":[[2021,8]]},"abstract":"<jats:p>Designing an incentive compatible auction that maximizes expected revenue is an intricate task. The single-item case was resolved in a seminal piece of work by Myerson in 1981. Even after 30--40 years of intense research, the problem remains unsolved for settings with two or more items. We overview recent research results that show how tools from deep learning are shaping up to become a powerful tool for the automated design of near-optimal auctions auctions. In this approach, an auction is modeled as a multilayer neural network, with optimal auction design framed as a constrained learning problem that can be addressed with standard machine learning pipelines. Through this approach, it is possible to recover to a high degree of accuracy essentially all known analytically derived solutions for multi-item settings and obtain novel mechanisms for settings in which the optimal mechanism is unknown.<\/jats:p>","DOI":"10.1145\/3470442","type":"journal-article","created":{"date-parts":[[2021,7,26]],"date-time":"2021-07-26T16:09:42Z","timestamp":1627315782000},"page":"109-116","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":28,"title":["Optimal auctions through deep learning"],"prefix":"10.1145","volume":"64","author":[{"given":"Paul","family":"D\u00fctting","sequence":"first","affiliation":[{"name":"Google Research, Z\u00fcrich, Switzerland"}]},{"given":"Zhe","family":"Feng","sequence":"additional","affiliation":[{"name":"Harvard University, Cambridge, MA"}]},{"given":"Harikrishna","family":"Narasimhan","sequence":"additional","affiliation":[{"name":"Google Research, Mountain View, CA"}]},{"given":"David C.","family":"Parkes","sequence":"additional","affiliation":[{"name":"Harvard University, MA"}]},{"given":"Sai S.","family":"Ravindranath","sequence":"additional","affiliation":[{"name":"Harvard University, Cambridge MA"}]}],"member":"320","published-online":{"date-parts":[[2021,7,26]]},"reference":[{"key":"e_1_2_1_1_1","doi-asserted-by":"publisher","DOI":"10.1109\/FOCS.2014.11"},{"key":"e_1_2_1_2_1","doi-asserted-by":"publisher","DOI":"10.1145\/2213977.2214021"},{"key":"e_1_2_1_3_1","first-page":"110","article-title":"Complexity of mechanism design. In Proceedings of the 18th Conference on Uncertainty","volume":"103","author":"Conitzer V.","year":"2002","unstructured":"Conitzer , V. , Sandholm , T . Complexity of mechanism design. In Proceedings of the 18th Conference on Uncertainty in Artificial Intelligence , 2002 , 103 -- 110 . Conitzer, V., Sandholm, T. Complexity of mechanism design. In Proceedings of the 18th Conference on Uncertainty in Artificial Intelligence, 2002, 103--110.","journal-title":"Artificial Intelligence"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.5555\/3495724.3496143"},{"key":"e_1_2_1_5_1","volume-title":"Proceedings of the 14th ACM Conference on Electronic Commerce","author":"Daskalakis C.","year":"2013","unstructured":"Daskalakis , C. , Deckelbaum , A. , Tzamos , C. Mechanism design via optimal transport . In Proceedings of the 14th ACM Conference on Electronic Commerce , 2013 , 269--286. Daskalakis, C., Deckelbaum, A., Tzamos, C. Mechanism design via optimal transport. In Proceedings of the 14th ACM Conference on Electronic Commerce, 2013, 269--286."},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.3982\/ECTA12618"},{"key":"e_1_2_1_7_1","volume-title":"Proceedings of the 36th International Conference on Machine Learning","author":"D\u00fctting P.","year":"2019","unstructured":"D\u00fctting , P. , Feng , Z. , Narasimhan , H. , Parkes , D.C. , Ravindranath , S.S. Optimal auctions through deep learning . In Proceedings of the 36th International Conference on Machine Learning , 2019 , 1706--1715. D\u00fctting, P., Feng, Z., Narasimhan, H., Parkes, D.C., Ravindranath, S.S. Optimal auctions through deep learning. In Proceedings of the 36th International Conference on Machine Learning, 2019, 1706--1715."},{"key":"e_1_2_1_8_1","first-page":"3","article-title":"Payment rules through discriminant-based classifiers","volume":"1","author":"D\u00fctting P.","year":"2014","unstructured":"D\u00fctting , P. , Fischer , F. , Jirapinyo , P. , Lai , J. , Lubin , B. , Parkes , D.C . Payment rules through discriminant-based classifiers . ACM Trans. Econ. Comput. 1 , 3 ( 2014 ), 5. D\u00fctting, P., Fischer, F., Jirapinyo, P., Lai, J., Lubin, B., Parkes, D.C. Payment rules through discriminant-based classifiers. ACM Trans. Econ. Comput. 1, 3 (2014), 5.","journal-title":"ACM Trans. Econ. Comput."},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.5555\/3237383.3237439"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1137\/16M1072218"},{"key":"e_1_2_1_11_1","volume-title":"Proceedings of the 13th International Conference on Artificial Intelligence and Statistics","author":"Glorot X.","year":"2010","unstructured":"Glorot , X. , Bengio , Y. Understanding the difficulty of training deep feedforward neural networks . In Proceedings of the 13th International Conference on Artificial Intelligence and Statistics , 2010 . Glorot, X., Bengio, Y. Understanding the difficulty of training deep feedforward neural networks. In Proceedings of the 13th International Conference on Artificial Intelligence and Statistics, 2010."},{"key":"e_1_2_1_12_1","doi-asserted-by":"publisher","DOI":"10.5555\/3304415.3304453"},{"key":"e_1_2_1_13_1","volume-title":"Proceedings of the 24th AAAI Conference on Artificial Intelligence","author":"Guo M.","year":"2010","unstructured":"Guo , M. , Conitzer , V. Computationally feasible automated mechanism design: General approach and case studies . In Proceedings of the 24th AAAI Conference on Artificial Intelligence , 2010 . Guo, M., Conitzer, V. Computationally feasible automated mechanism design: General approach and case studies. In Proceedings of the 24th AAAI Conference on Artificial Intelligence, 2010."},{"key":"e_1_2_1_14_1","volume-title":"Proceedings of the 29th Conference on Neural Information Processing Systems","author":"Hartford J.S.","year":"2016","unstructured":"Hartford , J.S. , Wright , J.R. , Leyton-Brown , K. Deep learning for predicting human strategic behavior . In Proceedings of the 29th Conference on Neural Information Processing Systems , 2016 , 2424--2432. Hartford, J.S., Wright, J.R., Leyton-Brown, K. Deep learning for predicting human strategic behavior. In Proceedings of the 29th Conference on Neural Information Processing Systems, 2016, 2424--2432."},{"key":"e_1_2_1_15_1","first-page":"127","article-title":"Bundling as an optimal selling mechanism for a multiple-good monopolist","volume":"1","author":"Manelli A.","year":"2006","unstructured":"Manelli , A. , Vincent , D . Bundling as an optimal selling mechanism for a multiple-good monopolist . J. Econ. Theory 1 , 127 ( 2006 ), 1--35. Manelli, A., Vincent, D. Bundling as an optimal selling mechanism for a multiple-good monopolist. J. Econ. Theory 1, 127 (2006), 1--35.","journal-title":"J. Econ. Theory"},{"key":"e_1_2_1_16_1","volume-title":"Proceedings of the 28th Conference on Neural Information Processing Systems","author":"Morgenstern J.","year":"2015","unstructured":"Morgenstern , J. , Roughgarden , T. On the pseudo-dimension of nearly optimal auctions . In Proceedings of the 28th Conference on Neural Information Processing Systems , 2015 , 136--144. Morgenstern, J., Roughgarden, T. On the pseudo-dimension of nearly optimal auctions. In Proceedings of the 28th Conference on Neural Information Processing Systems, 2015, 136--144."},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1287\/moor.6.1.58"},{"key":"e_1_2_1_18_1","volume-title":"Proceedings of the 25th International Joint Conference on Artificial Intelligence","author":"Narasimhan H.","year":"2016","unstructured":"Narasimhan , H. , Agarwal , S. , Parkes , D.C. Automated mechanism design without money via machine learning . In Proceedings of the 25th International Joint Conference on Artificial Intelligence , 2016 , 433--439. Narasimhan, H., Agarwal, S., Parkes, D.C. Automated mechanism design without money via machine learning. In Proceedings of the 25th International Joint Conference on Artificial Intelligence, 2016, 433--439."},{"key":"e_1_2_1_19_1","first-page":"51","article-title":"Bundling decisions by a multiproduct monopolist with incomplete information","volume":"2","author":"Palfrey T","year":"1983","unstructured":"Palfrey , T . Bundling decisions by a multiproduct monopolist with incomplete information . Econometrica 2 , 51 ( 1983 ), 463--483. Palfrey, T. Bundling decisions by a multiproduct monopolist with incomplete information. Econometrica 2, 51 (1983), 463--483.","journal-title":"Econometrica"},{"key":"e_1_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.2202\/1935-1704.1664"},{"key":"e_1_2_1_21_1","volume-title":"Proceedings of the 35th AAAI Conference on Artificial Intelligence","author":"Rahme J.","year":"2021","unstructured":"Rahme , J. , Jelassi , S. , Bruna , J. , Weinberg , S.M. A permutation-equivariant neural network architecture for auction design . In Proceedings of the 35th AAAI Conference on Artificial Intelligence ( 2021 ). Rahme, J., Jelassi, S., Bruna, J., Weinberg, S.M. A permutation-equivariant neural network architecture for auction design. In Proceedings of the 35th AAAI Conference on Artificial Intelligence (2021)."},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1016\/0304-4068(87)90007-3"},{"key":"e_1_2_1_23_1","first-page":"63","article-title":"Automated design of revenue-maximizing combinatorial auctions","volume":"5","author":"Sandholm T.","year":"2015","unstructured":"Sandholm , T. , Likhodedov , A . Automated design of revenue-maximizing combinatorial auctions . Oper. Res. 5 , 63 ( 2015 ), 1000--1025. Sandholm, T., Likhodedov, A. Automated design of revenue-maximizing combinatorial auctions. Oper. Res. 5, 63 (2015), 1000--1025.","journal-title":"Oper. Res."},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.5555\/3306127.3331696"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v31i1.10592"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1111\/j.1540-6261.1961.tb02789.x"}],"container-title":["Communications of the ACM"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3470442","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3470442","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T19:30:26Z","timestamp":1750188626000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3470442"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2021,7,26]]},"references-count":26,"journal-issue":{"issue":"8","published-print":{"date-parts":[[2021,8]]}},"alternative-id":["10.1145\/3470442"],"URL":"https:\/\/doi.org\/10.1145\/3470442","relation":{},"ISSN":["0001-0782","1557-7317"],"issn-type":[{"value":"0001-0782","type":"print"},{"value":"1557-7317","type":"electronic"}],"subject":[],"published":{"date-parts":[[2021,7,26]]},"assertion":[{"value":"2021-07-26","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}