{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,9]],"date-time":"2026-03-09T21:34:52Z","timestamp":1773092092935,"version":"3.50.1"},"reference-count":62,"publisher":"Association for Computing Machinery (ACM)","issue":"6","license":[{"start":{"date-parts":[[2020,9,18]],"date-time":"2020-09-18T00:00:00Z","timestamp":1600387200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"name":"GRF","award":["15210119"],"award-info":[{"award-number":["15210119"]}]},{"name":"CRF","award":["C5026-18G"],"award-info":[{"award-number":["C5026-18G"]}]},{"name":"ITF","award":["UICP\/UIM\/363"],"award-info":[{"award-number":["UICP\/UIM\/363"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Intell. Syst. Technol."],"published-print":{"date-parts":[[2020,12,31]]},"abstract":"<jats:p>In this article, we study solder paste inspection (SPI), an important stage that is used in the semiconductor manufacturing industry, where abnormal boards should be detected. A highly accurate SPI can substantially reduce human expert involvement, as well as reduce the waste in disposing of the boards in good condition. A key difference today is that because of increasing demand in board customization, the number of board types increases substantially and quantity of the boards produced in each type decreases. Thus, the previous approaches where a fine-tuned model is developed for each board type are no longer viable.<\/jats:p>\n          <jats:p>Intrinsically, our problem is an anomaly detection problem. A major specialty in today\u2019s SPI is that the target tasks for prediction cannot be fully pre-determined due to context changes during the solder paste printing stage. Our experiences show that a conventional approach to first define a set of tasks and train these tasks offline will lead to low accuracy. Here, we propose a novel multi-task approach, where the performance of all target tasks is ensured simultaneously. We note that the SPI process is streamlined and automatic, allowing the SPI time for only a few seconds. We propose a fast clustering algorithm that reuses existing models to avoid retraining and fine tune in the inference phase. We evaluate our approach using 3-month data collected from production lines. We show that we can reduce 81.28% of false alarms. This can translate to annual savings of $11.3 million.<\/jats:p>","DOI":"10.1145\/3383261","type":"journal-article","created":{"date-parts":[[2020,7,7]],"date-time":"2020-07-07T12:36:50Z","timestamp":1594125410000},"page":"1-17","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":13,"title":["Contextual Anomaly Detection in Solder Paste Inspection with Multi-Task Learning"],"prefix":"10.1145","volume":"11","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4342-6015","authenticated-orcid":false,"given":"Zimu","family":"Zheng","sequence":"first","affiliation":[{"name":"Hong Kong Polytechnic University and Huawei Cloud, Bantian, Longgang, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jie","family":"Pu","sequence":"additional","affiliation":[{"name":"Huawei Cloud, Bantian, Longgang, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Linghui","family":"Liu","sequence":"additional","affiliation":[{"name":"Huawei Cloud, Bantian, Longgang, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dan","family":"Wang","sequence":"additional","affiliation":[{"name":"Hong Kong Polytechnic University, Hung Hom, Hong Kong"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiangming","family":"Mei","sequence":"additional","affiliation":[{"name":"Huawei Cloud, Bantian, Longgang, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Sen","family":"Zhang","sequence":"additional","affiliation":[{"name":"Huawei Cloud, Bantian, Longgang, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7578-2738","authenticated-orcid":false,"given":"Quanyu","family":"Dai","sequence":"additional","affiliation":[{"name":"Hong Kong Polytechnic University, Hung Hom, Hong Kong"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2020,9,18]]},"reference":[{"key":"e_1_2_2_1_1","volume-title":"Retrieved","year":"2018"},{"key":"e_1_2_2_2_1","volume-title":"Retrieved","author":"Global PCB Co. Ltd.","year":"2018"},{"key":"e_1_2_2_3_1","volume-title":"Retrieved","author":"Riddle M. E.","year":"2019"},{"key":"e_1_2_2_4_1","first-page":"3","article-title":"Classification of solder joint using feature selection based on Bayes and support vector machine","volume":"3","author":"Wu H.","year":"2013","journal-title":"IEEE Transactions on Components, Packaging and Manufacturing Technology"},{"key":"e_1_2_2_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCPMT.2011.2168531"},{"key":"e_1_2_2_6_1","volume-title":"Proceedings of the 2007 IEEE International Symposium on Circuits and Systems. 1021--1024","author":"Acciani G."},{"key":"e_1_2_2_7_1","first-page":"5","article-title":"Feature-extraction-based inspection algorithm for IC solder joints","volume":"1","author":"Wu F.","year":"2011","journal-title":"IEEE Transactions on Components, Packaging and Manufacturing Technology"},{"key":"e_1_2_2_8_1","volume-title":"Proceedings of the 41st SICE Annual Conference (SICE\u201902)","volume":"4","author":"Ibrahim Z."},{"key":"e_1_2_2_9_1","doi-asserted-by":"publisher","DOI":"10.1109\/TII.2011.2166797"},{"key":"e_1_2_2_10_1","first-page":"1","article-title":"A new IC solder joint inspection method for an automatic optical inspection system based on an improved visual background extraction algorithm","volume":"6","author":"Cai N.","year":"2016","journal-title":"IEEE Transactions on Components, Packaging and Manufacturing Technology"},{"key":"e_1_2_2_11_1","first-page":"2","article-title":"IC solder joint inspection via robust principle component analysis","volume":"7","author":"Cai N.","year":"2017","journal-title":"IEEE Transactions on Components, Packaging and Manufacturing Technology"},{"key":"e_1_2_2_12_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCPMT.2018.2812815"},{"key":"e_1_2_2_13_1","doi-asserted-by":"publisher","DOI":"10.1145\/1541880.1541882"},{"key":"e_1_2_2_14_1","doi-asserted-by":"publisher","DOI":"10.1145\/335191.335388"},{"key":"e_1_2_2_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/376284.375668"},{"key":"e_1_2_2_16_1","doi-asserted-by":"publisher","DOI":"10.1145\/2983323.2983660"},{"key":"e_1_2_2_17_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2007.1009"},{"key":"e_1_2_2_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3097983.3098143"},{"key":"e_1_2_2_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/1835804.1835907"},{"key":"e_1_2_2_20_1","doi-asserted-by":"publisher","DOI":"10.1023\/A:1007379606734"},{"key":"e_1_2_2_21_1","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2009.191"},{"key":"e_1_2_2_22_1","unstructured":"Yu Zhang and Qiang Yang. 2017. A survey on multi-task learning. arXiv:1707.08114  Yu Zhang and Qiang Yang. 2017. A survey on multi-task learning. arXiv:1707.08114"},{"key":"e_1_2_2_23_1","volume-title":"Proceedings of the 2015 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP\u201915)","author":"Wu Z."},{"key":"e_1_2_2_24_1","volume-title":"Proceedings of 2013 IEEE Conference on Computer Vision and Pattern Recognition (CVPR\u201913)","author":"Yuan C."},{"key":"e_1_2_2_25_1","volume-title":"Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition (CVPR\u201909)","author":"Wang X."},{"key":"e_1_2_2_26_1","doi-asserted-by":"publisher","DOI":"10.1145\/2020408.2020549"},{"key":"e_1_2_2_27_1","volume-title":"Proceedings of the 31st AAAI Conference on Artificial Intelligence (AAAI\u201917)","author":"Zhang Yu","year":"2017"},{"key":"e_1_2_2_28_1","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2019\/615"},{"key":"e_1_2_2_29_1","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2537337"},{"key":"e_1_2_2_30_1","volume-title":"Proceedings of the 2016 International Joint Conference on Artificial Intelligence (IJCAI\u201916)","author":"Zhang Xianchao","year":"2016"},{"key":"e_1_2_2_31_1","volume-title":"Proceedings of the 2016 International Joint Conference on Artificial Intelligence (IJCAI\u201916)","author":"Isele David","year":"2016"},{"key":"e_1_2_2_32_1","volume-title":"Proceedings of the 28th AAAI Conference on Artificial Intelligence (AAAI\u201914)","author":"Han Lei","year":"2014"},{"key":"e_1_2_2_33_1","article-title":"Learning multiple tasks with kernel methods","author":"Evgeniou Theodoros","year":"2005","journal-title":"Journal of Machine Learning Research 6"},{"key":"e_1_2_2_34_1","unstructured":"Tsuyoshi Kato Hisashi Kashima Masashi Sugiyama and Kiyoshi Asai. 2008. Multi-task learning via conic programming. In Advances in Neural Information Processing Systems 20 (NIPS\u201908). 737--744.  Tsuyoshi Kato Hisashi Kashima Masashi Sugiyama and Kiyoshi Asai. 2008. Multi-task learning via conic programming. In Advances in Neural Information Processing Systems 20 (NIPS\u201908). 737--744."},{"key":"e_1_2_2_35_1","volume-title":"Proceedings of the 23rd AAAI Conference on Artificial Intelligence (AAAI\u201908)","volume":"1","author":"Larochelle Hugo","year":"2008"},{"key":"e_1_2_2_36_1","doi-asserted-by":"publisher","DOI":"10.1145\/3219819.3219878"},{"key":"e_1_2_2_37_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00391"},{"key":"e_1_2_2_38_1","doi-asserted-by":"publisher","DOI":"10.2307\/2528823"},{"key":"e_1_2_2_39_1","volume-title":"Numerical Ecology","author":"Legendre Pierre"},{"key":"e_1_2_2_40_1","doi-asserted-by":"publisher","DOI":"10.2307\/1224438"},{"key":"e_1_2_2_41_1","volume-title":"Retrieved","author":"Library Python Standard","year":"2018"},{"key":"e_1_2_2_42_1","volume-title":"Retrieved","author":"Learn Scikit","year":"2018"},{"key":"e_1_2_2_43_1","volume-title":"Retrieved","author":"Learn Scikit","year":"2018"},{"key":"e_1_2_2_44_1","volume-title":"Retrieved","year":"2018"},{"key":"e_1_2_2_45_1","volume-title":"Proceedings of the 2009 9th IEEE International Conference on Data Mining (ICDM\u201909)","author":"Gu Q."},{"key":"e_1_2_2_46_1","volume-title":"Retrieved","author":"Learn Scikit","year":"2018"},{"key":"e_1_2_2_47_1","volume-title":"Retrieved","year":"2018"},{"key":"e_1_2_2_48_1","volume-title":"Retrieved","author":"Learn Scikit","year":"2018"},{"key":"e_1_2_2_49_1","volume-title":"Retrieved","author":"Learn Scikit","year":"2018"},{"key":"e_1_2_2_50_1","volume-title":"The solder paste printing process: Critical parameters, defect scenarios, specifications, and cost reduction. Soldering 8 Surface Mount Technology 23, 4","author":"Huang Chien-Yi","year":"2011"},{"key":"e_1_2_2_51_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS.2019.00107"},{"key":"e_1_2_2_52_1","doi-asserted-by":"publisher","DOI":"10.1109\/TSUSC.2019.2932045"},{"key":"e_1_2_2_53_1","doi-asserted-by":"publisher","DOI":"10.1145\/3208903.3208913"},{"key":"e_1_2_2_54_1","volume-title":"Retrieved","year":"2018"},{"key":"e_1_2_2_55_1","volume-title":"Retrieved","author":"Riley Jenn","year":"2017"},{"key":"e_1_2_2_56_1","volume-title":"Proceedings of the 2016 International Joint Conference on Artificial Intelligence. 2965--2971","author":"\u00a0al Ruobing Xie","year":"2016"},{"key":"e_1_2_2_57_1","volume-title":"Retrieved","author":"BETS Research Group","year":"2017"},{"key":"e_1_2_2_58_1","doi-asserted-by":"publisher","DOI":"10.1145\/3276774.3276795"},{"key":"e_1_2_2_59_1","doi-asserted-by":"publisher","DOI":"10.1145\/2993422.2993577"},{"key":"e_1_2_2_60_1","volume-title":"Proceedings of the Statistics Canada Symposium. Catalogue no. 11-522-XIE.","author":"Dunstan Tim","year":"2005"},{"key":"e_1_2_2_61_1","doi-asserted-by":"crossref","unstructured":"Australian Institute of Health and Welfare. 2018. METeOR: Australian National Health Metadata Repository. Available at https:\/\/brickschema.org\/.  Australian Institute of Health and Welfare. 2018. METeOR: Australian National Health Metadata Repository. Available at https:\/\/brickschema.org\/.","DOI":"10.21820\/23987073.2018.2.80"},{"key":"e_1_2_2_62_1","volume-title":"Retrieved","author":"IEC JTC","year":"2018"}],"container-title":["ACM Transactions on Intelligent Systems and Technology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3383261","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3383261","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T22:02:02Z","timestamp":1750197722000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3383261"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,9,18]]},"references-count":62,"journal-issue":{"issue":"6","published-print":{"date-parts":[[2020,12,31]]}},"alternative-id":["10.1145\/3383261"],"URL":"https:\/\/doi.org\/10.1145\/3383261","relation":{},"ISSN":["2157-6904","2157-6912"],"issn-type":[{"value":"2157-6904","type":"print"},{"value":"2157-6912","type":"electronic"}],"subject":[],"published":{"date-parts":[[2020,9,18]]},"assertion":[{"value":"2019-07-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-02-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2020-09-18","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}