{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T04:22:58Z","timestamp":1750220578198,"version":"3.41.0"},"reference-count":31,"publisher":"Association for Computing Machinery (ACM)","issue":"1","license":[{"start":{"date-parts":[[2020,9,4]],"date-time":"2020-09-04T00:00:00Z","timestamp":1599177600000},"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":["SIGMOD Rec."],"published-print":{"date-parts":[[2020,9,4]]},"abstract":"<jats:p>Deep Convolutional Neural Networks (CNNs) now match human accuracy in many image prediction tasks, resulting in a growing adoption in e-commerce, radiology, and other domains. Naturally, \"explaining\" CNN predictions is a key concern for many users. Since the internal workings of CNNs are unintuitive for most users, occlusion-based explanations (OBE) are popular for understanding which parts of an image matter most for a prediction. One occludes a region of the image using a patch and moves it around to produce a heatmap of changes to the prediction probability. This approach is computationally expensive due to the large number of re-inference requests produced, which wastes time and raises resource costs. We tackle this issue by casting the OBE task as a new instance of the classical incremental view maintenance problem. We create a novel and comprehensive algebraic framework for incremental CNN inference combining materialized views with multi-query optimization to reduce computational costs. We then present two novel approximate inference optimizations that exploit the semantics of CNNs and the OBE task to further reduce runtimes. We prototype our ideas in a tool we call Krypton. Experiments with real data and CNNs show that Krypton reduces runtimes by up to 5x (resp. 35x) to produce exact (resp. high-quality approximate) results without raising resource requirements.<\/jats:p>","DOI":"10.1145\/3422648.3422663","type":"journal-article","created":{"date-parts":[[2020,9,4]],"date-time":"2020-09-04T10:09:28Z","timestamp":1599214168000},"page":"61-68","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":3,"title":["Query Optimization for Faster Deep CNN Explanations"],"prefix":"10.1145","volume":"49","author":[{"given":"Supun","family":"Nakandala","sequence":"first","affiliation":[{"name":"University of California, San Diego, CA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Arun","family":"Kumar","sequence":"additional","affiliation":[{"name":"University of California, San Diego, CA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yannis","family":"Papakonstantinou","sequence":"additional","affiliation":[{"name":"University of California, San Diego, CA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2020,9,4]]},"reference":[{"key":"e_1_2_1_1_1","volume-title":"https:\/\/bit.ly\/36300H9. Accessed","author":"Detecting Diabetic Retinopathy Earns AI","year":"2020","unstructured":"AI Device for Detecting Diabetic Retinopathy Earns Swift FDA Approval . https:\/\/bit.ly\/36300H9. Accessed April 30, 2020 . AI Device for Detecting Diabetic Retinopathy Earns Swift FDA Approval. https:\/\/bit.ly\/36300H9. Accessed April 30, 2020."},{"key":"e_1_2_1_2_1","doi-asserted-by":"publisher","DOI":"10.5555\/3348507"},{"key":"e_1_2_1_3_1","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA.2018.00051"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1145\/3131885.3131906"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.14778\/3137628.3137633"},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.5555\/2480856"},{"key":"e_1_2_1_7_1","volume":"201","author":"de Vries S. E.","unstructured":"S. E. de Vries The Projective Field of a Retinal Amacrine Cell. In Journal of Neuroscience , 201 1. S. E. de Vries et al. The Projective Field of a Retinal Amacrine Cell. In Journal of Neuroscience, 2011.","journal-title":"The Projective Field of a Retinal Amacrine Cell. In Journal of Neuroscience"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.5555\/645927.672356"},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.5555\/3086952"},{"key":"e_1_2_1_10_1","volume":"201","author":"Jung K.-H.","unstructured":"K.-H. Jung Deep Learning for Medical Image Analysis: Applications to Computed Tomography and Magnetic Resonance Imaging. Hanyang Medical Reviews , 201 7. K.-H. Jung et al. Deep Learning for Medical Image Analysis: Applications to Computed Tomography and Magnetic Resonance Imaging. Hanyang Medical Reviews, 2017.","journal-title":"Hanyang Medical Reviews"},{"key":"e_1_2_1_11_1","volume-title":"Cell","author":"Kermany D. S.","year":"2018","unstructured":"D. S. Kermany Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning . Cell , 2018 . D. S. Kermany et al. Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning. Cell, 2018."},{"key":"e_1_2_1_12_1","doi-asserted-by":"publisher","DOI":"10.14778\/2536274.2536302"},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.1145\/2723372.2723713"},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1145\/2935694.2935698"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/3035918.3054775"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1145\/212433.220198"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/3299869.3319878"},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3329486.3329496"},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1145\/3299869.3319874"},{"key":"e_1_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.1145\/3299869.3319874"},{"key":"e_1_2_1_21_1","doi-asserted-by":"publisher","DOI":"10.1145\/3397461"},{"key":"e_1_2_1_22_1","doi-asserted-by":"publisher","DOI":"10.1145\/3318464.3389709"},{"key":"e_1_2_1_23_1","doi-asserted-by":"publisher","DOI":"10.1145\/2588555.2610519"},{"key":"e_1_2_1_24_1","doi-asserted-by":"publisher","DOI":"10.14778\/3352063.3352093"},{"key":"e_1_2_1_25_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.1145\/42201.42203"},{"key":"e_1_2_1_27_1","doi-asserted-by":"publisher","DOI":"10.5555\/3152676"},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2003.819861"},{"key":"e_1_2_1_29_1","volume-title":"Visualizing and Understanding Convolutional Networks. In European Conference on Computer Vision","author":"Zeiler M. D.","year":"2014","unstructured":"M. D. Zeiler and R. Fergus . Visualizing and Understanding Convolutional Networks. In European Conference on Computer Vision , 2014 . M. D. Zeiler and R. Fergus. Visualizing and Understanding Convolutional Networks. In European Conference on Computer Vision, 2014."},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1145\/2588555.2593678"},{"key":"e_1_2_1_31_1","volume-title":"ICLR","author":"Zintgraf L. M.","year":"2017","unstructured":"L. M. Zintgraf Visualizing Deep Neural Network Decisions: Prediction Difference Analysis . In ICLR , 2017 . L. M. Zintgraf et al. Visualizing Deep Neural Network Decisions: Prediction Difference Analysis. In ICLR, 2017."}],"container-title":["ACM SIGMOD Record"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3422648.3422663","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3422648.3422663","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T21:31:29Z","timestamp":1750195889000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3422648.3422663"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2020,9,4]]},"references-count":31,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2020,9,4]]}},"alternative-id":["10.1145\/3422648.3422663"],"URL":"https:\/\/doi.org\/10.1145\/3422648.3422663","relation":{},"ISSN":["0163-5808"],"issn-type":[{"type":"print","value":"0163-5808"}],"subject":[],"published":{"date-parts":[[2020,9,4]]},"assertion":[{"value":"2020-09-04","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}