{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T01:45:33Z","timestamp":1787017533305,"version":"build-2736575974"},"reference-count":93,"publisher":"Association for Computing Machinery (ACM)","issue":"1","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. ACM Manag. Data"],"published-print":{"date-parts":[[2026,4,2]]},"abstract":"<jats:p>\n                    Real-world AI\/ML workflows often apply inference computations to feature vectors joined from multiple datasets. To avoid the redundant AI\/ML computations caused by repeated data records in the join's output,\n                    <jats:italic toggle=\"yes\">factorized<\/jats:italic>\n                    ML has been proposed to decompose ML computations into sub-computations to be executed on each normalized dataset. However, there is insufficient discussion on how\n                    <jats:italic toggle=\"yes\">factorized<\/jats:italic>\n                    ML could impact AI\/ML inference over multi-way joins. To address the limitations, we propose a novel declarative InferF system, focusing on the factorization of arbitrary inference workflows represented as analyzable expressions over the multi-way joins. We formalize our problem to flexibly push down partial factorized computations to qualified nodes in the join tree to minimize the overall inference computation and join costs and propose two algorithms to resolve the problem: (1) a greedy algorithm based on a per-node cost function that estimates the influence on overall latency if a subset of factorized computations is pushed to a node, and (2) a genetic algorithm for iteratively enumerating and evaluating promising factorization plans. We implement InferF on Velox, an open-sourced database engine from Meta, evaluate it on real-world datasets, observed up to 11.3\u00d7 speedups, and systematically summarized the factors that determine when\n                    <jats:italic toggle=\"yes\">factorized<\/jats:italic>\n                    ML can benefit AI\/ML inference workflows.\n                  <\/jats:p>","DOI":"10.1145\/3786662","type":"journal-article","created":{"date-parts":[[2026,4,7]],"date-time":"2026-04-07T17:54:13Z","timestamp":1775584453000},"page":"1-28","source":"Crossref","is-referenced-by-count":1,"title":["InferF: Declarative Factorization of AI\/ML Inferences over Joins"],"prefix":"10.1145","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-6528-9324","authenticated-orcid":false,"given":"Kanchan","family":"Chowdhury","sequence":"first","affiliation":[{"name":"Arizona State University, Tempe, Arizona, USA and Marquette University, Milwaukee, Wisconsin, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8481-3292","authenticated-orcid":false,"given":"Lixi","family":"Zhou","sequence":"additional","affiliation":[{"name":"Arizona State University, Tempe, Arizona, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8902-222X","authenticated-orcid":false,"given":"Lulu","family":"Xie","sequence":"additional","affiliation":[{"name":"Arizona State University, Tempe, Arizona, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-7822-5450","authenticated-orcid":false,"given":"Xinwei","family":"Fu","sequence":"additional","affiliation":[{"name":"Amazon, Santa Clara, California, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9849-0711","authenticated-orcid":false,"given":"Jia","family":"Zou","sequence":"additional","affiliation":[{"name":"Arizona State University, Tempe, Arizona, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,4,7]]},"reference":[{"key":"e_1_2_1_1_1","unstructured":"[n.d.]. The total cost of ownership (tco) of amazon sagemaker. ([n.d.]). https:\/\/pages.awscloud.com\/NAMER-ln-GC-400-machine-learning-sagemaker-tco-learn-ty.html."},{"key":"e_1_2_1_2_1","unstructured":"2021. AWS re:Invent 2018 keynote. https:\/\/www.youtube.com\/watch?v=ZOIkOnW640A&t=5316s."},{"key":"e_1_2_1_3_1","unstructured":"2023. Amazon SageMaker Feature Store. https:\/\/aws.amazon.com\/sagemaker\/feature-store\/."},{"key":"e_1_2_1_4_1","unstructured":"2023. postgresML. https:\/\/postgresml.org\/docs\/guides\/transformers\/setup."},{"key":"e_1_2_1_5_1","volume-title":"TensorFlow Decision Forests (TF-DF). (Last accessed","author":"Last","year":"2023","unstructured":"Last accessed in Sept 2023. TensorFlow Decision Forests (TF-DF). (Last accessed in Sept 2023). https:\/\/github.com\/tensorflow\/decision-forests"},{"key":"e_1_2_1_6_1","unstructured":"Postgresql Documentation 02. 2025. 2025. Chapter 68.1. Row Estimation Examples. https:\/\/www.postgresql.org\/docs\/current\/row-estimation-examples.html."},{"key":"e_1_2_1_7_1","doi-asserted-by":"publisher","DOI":"10.1145\/3294052.3319694"},{"key":"e_1_2_1_8_1","volume-title":"Brainwash: A Data System for Feature Engineering. In Cidr.","author":"Anderson Michael R","year":"2013","unstructured":"Michael R Anderson, Dolan Antenucci, Victor Bittorf, Matthew Burgess, Michael J Cafarella, Arun Kumar, Feng Niu, Yongjoo Park, Christopher R\u00e9, and Ce Zhang. 2013. Brainwash: A Data System for Feature Engineering. In Cidr."},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1145\/3514221.3526045"},{"key":"e_1_2_1_10_1","volume-title":"ONNX: Open Neural Network Exchange. https:\/\/github.com\/onnx\/onnx.","author":"Bai Junjie","year":"2019","unstructured":"Junjie Bai, Fang Lu, Ke Zhang, et al. 2019. ONNX: Open Neural Network Exchange. https:\/\/github.com\/onnx\/onnx."},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1145\/3183713.3190662"},{"key":"e_1_2_1_12_1","unstructured":"Kristin P Bennett Michael C Ferris and Yannis E Ioannidis. 1991. A genetic algorithm for database query optimization. Technical Report. University of Wisconsin-Madison Department of Computer Sciences."},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.5555\/3379393"},{"key":"e_1_2_1_14_1","unstructured":"Matthias Boehm Iulian Antonov Sebastian Baunsgaard Mark Dokter Robert Ginth\u00f6r Kevin Innerebner Florijan Klezin Stefanie Lindstaedt Arnab Phani Benjamin Rath et al. 2019. SystemDS: A declarative machine learning system for the end-to-end data science lifecycle. arXiv preprint arXiv:1909.02976 (2019)."},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.14778\/3007263.3007279"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.14778\/3137628.3137633"},{"key":"e_1_2_1_17_1","volume-title":"14th USENIX Symposium on Networked Systems Design and Implementation (NSDI 17)","author":"Crankshaw Daniel","year":"2017","unstructured":"Daniel Crankshaw, Xin Wang, Guilio Zhou, Michael J Franklin, Joseph E Gonzalez, and Ion Stoica. 2017. Clipper: A {Low-Latency} online prediction serving system. In 14th USENIX Symposium on Networked Systems Design and Implementation (NSDI 17). 613--627."},{"key":"e_1_2_1_18_1","volume-title":"Proc. of the 11th Workshop on Mining and Learning with Graphs","author":"Dhurandhar Amit","year":"2013","unstructured":"Amit Dhurandhar. 2013. Auto-correlation dependent bounds for relational data. In Proc. of the 11th Workshop on Mining and Learning with Graphs. Chicago."},{"key":"e_1_2_1_19_1","unstructured":"EvaDB. 2023. EvaDB: Database system for AI-powered apps. Technical Report. https:\/\/github.com\/georgia-tech-db\/evadb"},{"key":"e_1_2_1_20_1","unstructured":"The PyTorch Foundation. 2018. LibTorch: PyTorch C API. https:\/\/pytorch.org\/cppdocs\/"},{"key":"e_1_2_1_21_1","unstructured":"Hector Garcia-Molina. 2008. Database systems: the complete book. Pearson Education India."},{"key":"e_1_2_1_22_1","volume-title":"Relational data mining","author":"Getoor Lise","unstructured":"Lise Getoor, Nir Friedman, Daphne Koller, and Avi Pfeffer. 2001. Learning probabilistic relational models. In Relational data mining. Springer, 307--335."},{"key":"e_1_2_1_23_1","doi-asserted-by":"crossref","unstructured":"Lise Getoor and Ben Taskar. 2007. Introduction to statistical relational learning.","DOI":"10.7551\/mitpress\/7432.001.0001"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.1145\/3620678.3624656"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1145\/3725340"},{"key":"e_1_2_1_26_1","unstructured":"Ga\u00ebl Guennebaud Beno\u00eet Jacob et al. 2010. Eigen v3. http:\/\/eigen.tuxfamily.org."},{"key":"e_1_2_1_27_1","volume-title":"Yggdrasil Decision Forests: A Fast and Extensible Decision Forests Library. arXiv preprint arXiv:2212.02934","author":"Guillame-Bert Mathieu","year":"2022","unstructured":"Mathieu Guillame-Bert, Sebastian Bruch, Richard Stotz, and Jan Pfeifer. 2022. Yggdrasil Decision Forests: A Fast and Extensible Decision Forests Library. arXiv preprint arXiv:2212.02934 (2022)."},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jda.2005.07.005"},{"key":"e_1_2_1_29_1","doi-asserted-by":"publisher","DOI":"10.14778\/2367502.2367510"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.21105\/joss.00097"},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.14778\/3611479.3611509"},{"key":"e_1_2_1_32_1","doi-asserted-by":"publisher","DOI":"10.1145\/1273496.1273543"},{"key":"e_1_2_1_33_1","doi-asserted-by":"publisher","DOI":"10.14778\/3317315.3317323"},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.1145\/3422648.3422659"},{"key":"e_1_2_1_35_1","first-page":"259","article-title":"Linkage and autocorrelation cause feature selection bias in relational learning","volume":"2","author":"Jensen David","year":"2002","unstructured":"David Jensen and Jennifer Neville. 2002. Linkage and autocorrelation cause feature selection bias in relational learning. In ICML, Vol. 2. 259--266.","journal-title":"ICML"},{"key":"e_1_2_1_36_1","doi-asserted-by":"publisher","DOI":"10.14778\/3611479.3611539"},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.14778\/3551793.3551801"},{"key":"e_1_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2010.57"},{"key":"e_1_2_1_39_1","doi-asserted-by":"publisher","DOI":"10.1145\/3595360.3595858"},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1145\/3209889.3209896"},{"key":"e_1_2_1_41_1","volume-title":"Jia Li, Chenghao Mou, Carlos Mu\u00f1oz Ferrandis, Yacine Jernite, Margaret Mitchell, Sean Hughes, Thomas Wolf, Dzmitry Bahdanau, Leandro von Werra, and Harm de Vries.","author":"Kocetkov Denis","year":"2022","unstructured":"Denis Kocetkov, Raymond Li, Loubna Ben Allal, Jia Li, Chenghao Mou, Carlos Mu\u00f1oz Ferrandis, Yacine Jernite, Margaret Mitchell, Sean Hughes, Thomas Wolf, Dzmitry Bahdanau, Leandro von Werra, and Harm de Vries. 2022. The Stack: 3 TB of permissively licensed source code. Preprint (2022)."},{"key":"e_1_2_1_42_1","volume-title":"Propositionalization approaches to relational data mining. Relational data mining","author":"Kramer Stefan","year":"2001","unstructured":"Stefan Kramer, Nada Lavrac, and Peter Flach. 2001. Propositionalization approaches to relational data mining. Relational data mining (2001), 262--291."},{"key":"e_1_2_1_43_1","doi-asserted-by":"publisher","DOI":"10.1145\/2723372.2723713"},{"key":"e_1_2_1_44_1","doi-asserted-by":"publisher","DOI":"10.1145\/2882903.2882952"},{"key":"e_1_2_1_45_1","doi-asserted-by":"publisher","DOI":"10.14778\/3342263.3342633"},{"key":"e_1_2_1_46_1","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/P15-1065"},{"key":"e_1_2_1_47_1","doi-asserted-by":"publisher","DOI":"10.14778\/2850583.2850594"},{"key":"e_1_2_1_48_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE60146.2024.00391"},{"key":"e_1_2_1_49_1","doi-asserted-by":"publisher","DOI":"10.1145\/3299869.3319878"},{"key":"e_1_2_1_50_1","doi-asserted-by":"publisher","DOI":"10.14778\/3632093.3632095"},{"key":"e_1_2_1_51_1","doi-asserted-by":"publisher","DOI":"10.1145\/2766462.2767733"},{"key":"e_1_2_1_52_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-71273-4_36"},{"key":"e_1_2_1_53_1","doi-asserted-by":"publisher","DOI":"10.1109\/BigData.2016.7840826"},{"key":"e_1_2_1_54_1","doi-asserted-by":"publisher","DOI":"10.5555\/1182635.1164217"},{"key":"e_1_2_1_55_1","volume-title":"Workshop SRL, Int. Joint. Conf. on AI","volume":"25","author":"Neville Jennifer","year":"2003","unstructured":"Jennifer Neville, Matthew Rattigan, and David Jensen. 2003. Statistical relational learning: Four claims and a survey. In Workshop SRL, Int. Joint. Conf. on AI, Vol. 25."},{"key":"e_1_2_1_56_1","doi-asserted-by":"publisher","DOI":"10.14778\/3007263.3007312"},{"key":"e_1_2_1_57_1","doi-asserted-by":"publisher","unstructured":"Dan Olteanu and Maximilian Schleich. 2016. Factorized Databases. 45 2 (Sept. 2016) 5--16. doi:10.1145\/3003665.3003667","DOI":"10.1145\/3003665.3003667"},{"key":"e_1_2_1_58_1","doi-asserted-by":"publisher","DOI":"10.14778\/3213880.3213890"},{"key":"e_1_2_1_59_1","doi-asserted-by":"publisher","DOI":"10.1145\/3514221.3526141"},{"key":"e_1_2_1_60_1","doi-asserted-by":"publisher","DOI":"10.1145\/3514221.3526141"},{"key":"e_1_2_1_61_1","first-page":"2825","article-title":"Scikit-learn: Machine learning in Python","author":"Pedregosa Fabian","year":"2011","unstructured":"Fabian Pedregosa, Ga\u00ebl Varoquaux, Alexandre Gramfort, Vincent Michel, Bertrand Thirion, Olivier Grisel, Mathieu Blondel, Peter Prettenhofer, Ron Weiss, Vincent Dubourg, et al. 2011. Scikit-learn: Machine learning in Python. Journal of machine learning research 12, Oct (2011), 2825--2830.","journal-title":"Journal of machine learning research 12"},{"key":"e_1_2_1_62_1","doi-asserted-by":"publisher","DOI":"10.14778\/3554821.3554829"},{"key":"e_1_2_1_63_1","doi-asserted-by":"publisher","DOI":"10.1145\/3448016.3457275"},{"key":"e_1_2_1_64_1","unstructured":"Project Hamlet. 2021. Project Hamlet. https:\/\/adalabucsd.github.io\/hamlet.html. Accessed: 2025-09-04."},{"key":"e_1_2_1_65_1","unstructured":"Lyle Ramshaw and Robert E Tarjan. 2012. On minimum-cost assignments in unbalanced bipartite graphs. HP Labs Palo Alto CA USA Tech. Rep. HPL-2012--40R1 20 (2012) 14."},{"key":"e_1_2_1_66_1","volume-title":"Markov logic networks. Machine learning 62, 1","author":"Richardson Matthew","year":"2006","unstructured":"Matthew Richardson and Pedro Domingos. 2006. Markov logic networks. Machine learning 62, 1 (2006), 107--136."},{"key":"e_1_2_1_67_1","volume-title":"https:\/\/www.postgresql.org\/docs\/current\/planner-optimizer.html. Postgres Professional","author":"Rogov Egor","year":"2023","unstructured":"Egor Rogov. 2023. PostgreSQL 14 Internals. https:\/\/www.postgresql.org\/docs\/current\/planner-optimizer.html. Postgres Professional (2023)."},{"key":"e_1_2_1_68_1","volume-title":"17th International Joint Conference on Artificial Intelligence","author":"Roth Dan","year":"2001","unstructured":"Dan Roth and Wen-tau Yih. 2001. Propositionalization of relational learning: An information extraction case study. In 17th International Joint Conference on Artificial Intelligence, Seattle."},{"key":"e_1_2_1_69_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cpc.2009.09.018"},{"key":"e_1_2_1_70_1","doi-asserted-by":"publisher","DOI":"10.1145\/3299869.3324961"},{"key":"e_1_2_1_71_1","doi-asserted-by":"publisher","DOI":"10.1145\/2882903.2882939"},{"key":"e_1_2_1_72_1","doi-asserted-by":"publisher","DOI":"10.1145\/3527333"},{"key":"e_1_2_1_73_1","doi-asserted-by":"publisher","DOI":"10.1145\/6314.6315"},{"key":"e_1_2_1_74_1","doi-asserted-by":"publisher","DOI":"10.1145\/3341301.3359658"},{"key":"e_1_2_1_75_1","volume-title":"25th International Conference on Machine Learning (ICML2008)","volume":"10","author":"Sonnenburg Soeren","year":"2008","unstructured":"Soeren Sonnenburg, Vojtech Franc, Elad Yom-Tov, and Michele Sebag. 2008. Pascal large scale learning challenge. In 25th International Conference on Machine Learning (ICML2008) Workshop. http:\/\/largescale. first. fraunhofer. de. J. Mach. Learn. Res, Vol. 10. 1937--1953."},{"key":"e_1_2_1_76_1","unstructured":"Apache Spark. 2024. PySpark. https:\/\/spark.apache.org\/docs\/latest\/api\/python\/index.html. Accessed: 2025-01-03."},{"key":"e_1_2_1_77_1","doi-asserted-by":"publisher","DOI":"10.1007\/s007780050040"},{"key":"e_1_2_1_78_1","doi-asserted-by":"publisher","DOI":"10.14778\/3275366.3284963"},{"key":"e_1_2_1_79_1","doi-asserted-by":"publisher","DOI":"10.14778\/3551793.3551847"},{"key":"e_1_2_1_80_1","doi-asserted-by":"publisher","DOI":"10.14778\/3704965.3704971"},{"key":"e_1_2_1_81_1","doi-asserted-by":"publisher","DOI":"10.14778\/3457390.3457399"},{"key":"e_1_2_1_82_1","doi-asserted-by":"publisher","DOI":"10.14778\/2735508.2735510"},{"key":"e_1_2_1_83_1","doi-asserted-by":"publisher","DOI":"10.1145\/3626246.3654754"},{"key":"e_1_2_1_84_1","doi-asserted-by":"publisher","DOI":"10.1145\/3725326"},{"key":"e_1_2_1_85_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICDE60146.2024.00008"},{"key":"e_1_2_1_86_1","doi-asserted-by":"publisher","DOI":"10.14778\/3547305.3547325"},{"key":"e_1_2_1_87_1","doi-asserted-by":"publisher","DOI":"10.48786\/EDBT.2024.61"},{"key":"e_1_2_1_88_1","doi-asserted-by":"publisher","DOI":"10.14778\/3611540.3611550"},{"key":"e_1_2_1_89_1","doi-asserted-by":"publisher","DOI":"10.14778\/3090163.3090167"},{"key":"e_1_2_1_90_1","doi-asserted-by":"publisher","DOI":"10.1145\/3183713.3196933"},{"key":"e_1_2_1_91_1","doi-asserted-by":"publisher","DOI":"10.14778\/3457390.3457392"},{"key":"e_1_2_1_92_1","volume-title":"Pangea: Monolithic Distributed Storage for Data Analytics. Proceedings of the VLDB Endowment 12","author":"Zou Jia","unstructured":"Jia Zou, Arun Iyengar, and Chris Jermaine. [n.d.]. Pangea: Monolithic Distributed Storage for Data Analytics. Proceedings of the VLDB Endowment 12, 6 ([n.d.])."},{"key":"e_1_2_1_93_1","doi-asserted-by":"publisher","DOI":"10.1007\/s00778-020-00605-w"}],"container-title":["Proceedings of the ACM on Management of Data"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3786662","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,7]],"date-time":"2026-04-07T20:03:09Z","timestamp":1775592189000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3786662"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4,2]]},"references-count":93,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,4,2]]}},"alternative-id":["10.1145\/3786662"],"URL":"https:\/\/doi.org\/10.1145\/3786662","relation":{},"ISSN":["2836-6573"],"issn-type":[{"value":"2836-6573","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,4,2]]}}}