{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T14:03:22Z","timestamp":1789999402661,"version":"4.0.1"},"reference-count":47,"publisher":"Association for Computing Machinery (ACM)","issue":"4","license":[{"start":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T00:00:00Z","timestamp":1789948800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"National Science Foundation","award":["Grant Nos. CNS-2328972, CCF-2324937, CNS-2122320, and CNS-2038960"],"award-info":[{"award-number":["Grant Nos. CNS-2328972, CCF-2324937, CNS-2122320, and CNS-2038960"]}]},{"name":"TUM Institute for Advanced Study in Germany; and Dieter Schwarz Courageous Research Grant"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Cyber-Phys. Syst."],"published-print":{"date-parts":[[2026,10,31]]},"abstract":"<jats:p>Traditional neural architecture pruning methods primarily focus on tradeoffs between inference accuracy and computational cost (e.g., FLOPs), without accounting for system-level behaviors. In the context of cyber-physical systems (CPSs), perception models and the downstream controllers that rely on their outputs are typically designed independently. In this article, we study the joint design of neural architecture pruning strategies and the controllers that utilize the resulting inference outputs. We propose two key innovations: (i) incorporating system-level safety metrics into the pruning optimization loop, and (ii) a regression-based scaling law method that efficiently models the relationship between pruning ratios and system-level safety outcomes. This enables rapid pruning guided by system-level constraints, avoiding the inefficiency of exhaustive search or conventional neural architecture search approaches that focus on local accuracy or latency metrics. We demonstrate our approach on YOLO-based perception models deployed in control systems for distance estimation. To the best of our knowledge, this is the first work to introduce a scaling-law-guided pruning method for YOLO-based perception models, offering a principled design automation framework for safety-aware, machine learning-enabled CPS.<\/jats:p>","DOI":"10.1145\/3828657","type":"journal-article","created":{"date-parts":[[2026,7,3]],"date-time":"2026-07-03T14:28:48Z","timestamp":1783088928000},"page":"1-25","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Controller-aware Perception Neural Network Pruning for CPS Safety Optimization"],"prefix":"10.1145","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-0820-6613","authenticated-orcid":false,"given":"Jingxian","family":"Chen","sequence":"first","affiliation":[{"name":"Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, Pennsylvania, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0634-1941","authenticated-orcid":false,"given":"Yukai","family":"Song","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, Pennsylvania, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1784-1386","authenticated-orcid":false,"given":"Shengjie","family":"Xu","sequence":"additional","affiliation":[{"name":"Department of Computer Science, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0162-9428","authenticated-orcid":false,"given":"Kyle","family":"Kessler","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, Pennsylvania, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-3688-1399","authenticated-orcid":false,"given":"Prateek","family":"Ganguli","sequence":"additional","affiliation":[{"name":"Department of Computer Science, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-3006-7447","authenticated-orcid":false,"given":"Arkaprava","family":"Gupta","sequence":"additional","affiliation":[{"name":"Department of Computer Science, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1371-2803","authenticated-orcid":false,"given":"Bineet","family":"Ghosh","sequence":"additional","affiliation":[{"name":"Department of Computer Science, The University of Alabama at Birmingham, Birmingham, Alabama, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8871-0298","authenticated-orcid":false,"given":"Parasara Sridhar","family":"Duggirala","sequence":"additional","affiliation":[{"name":"Department of Computer Science, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7774-8197","authenticated-orcid":false,"given":"Tianlong","family":"Chen","sequence":"additional","affiliation":[{"name":"Department of Computer Science, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0503-6235","authenticated-orcid":false,"given":"Samarjit","family":"Chakraborty","sequence":"additional","affiliation":[{"name":"Department of Computer Science, The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA and TUM Institute for Advanced Study, Garching, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4029-4034","authenticated-orcid":false,"given":"Jingtong","family":"Hu","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, University of Pittsburgh, Pittsburgh, Pennsylvania, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,9,21]]},"reference":[{"key":"e_1_3_1_2_2","doi-asserted-by":"publisher","DOI":"10.1109\/TRO.2014.2312453"},{"key":"e_1_3_1_3_2","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-control-071420-081941"},{"key":"e_1_3_1_4_2","doi-asserted-by":"publisher","DOI":"10.1109\/CDC.2008.4738704"},{"key":"e_1_3_1_5_2","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2016.2638961"},{"key":"e_1_3_1_6_2","doi-asserted-by":"publisher","DOI":"10.1109\/CDC.2017.8263977"},{"key":"e_1_3_1_7_2","doi-asserted-by":"publisher","DOI":"10.1145\/3302504.3311804"},{"key":"e_1_3_1_8_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2024.3480976"},{"issue":"4","key":"e_1_3_1_9_2","doi-asserted-by":"crossref","first-page":"249","DOI":"10.1561\/1000000045","article-title":"Resource-aware automotive control systems design: A cyber-physical systems approach","volume":"10","author":"Chang Wanli","year":"2016","unstructured":"Wanli Chang and Samarjit Chakraborty. 2016. Resource-aware automotive control systems design: A cyber-physical systems approach. Foundations and Trends\u00ae in Electronic Design Automation 10, 4 (2016), 249\u2013369.","journal-title":"Foundations and Trends\u00ae in Electronic Design Automation"},{"key":"e_1_3_1_10_2","unstructured":"Xiaodong Chen Yuxuan Hu Xiaokang Zhang Yanling Wang Cuiping Li Hong Chen and Jing Zhang. 2024. P \\({}^{2}\\) law: Scaling law for post-training after model pruning. arXiv:2411.10272. Retrieved from https:\/\/arxiv.org\/abs\/2411.10272"},{"key":"e_1_3_1_11_2","doi-asserted-by":"publisher","DOI":"10.1109\/TIV.2023.3318070"},{"key":"e_1_3_1_12_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-15297-9_9"},{"key":"e_1_3_1_13_2","doi-asserted-by":"publisher","DOI":"10.1061\/JTEPBS.TEENG-8446"},{"key":"e_1_3_1_14_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01544"},{"key":"e_1_3_1_15_2","doi-asserted-by":"publisher","DOI":"10.1145\/2463209.2488855"},{"key":"e_1_3_1_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/RTCSA55878.2022.00008"},{"key":"e_1_3_1_17_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-540-31954-2_19"},{"key":"e_1_3_1_18_2","doi-asserted-by":"publisher","DOI":"10.1007\/11730637_21"},{"key":"e_1_3_1_19_2","doi-asserted-by":"publisher","DOI":"10.1109\/ASPDAC.2011.5722188"},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCST.2014.2301795"},{"key":"e_1_3_1_21_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2022.3198905"},{"key":"e_1_3_1_22_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCPS61052.2024.00016"},{"key":"e_1_3_1_23_2","doi-asserted-by":"crossref","unstructured":"Jordan Hoffmann Sebastian Borgeaud Arthur Mensch Elena Buchatskaya Trevor Cai Eliza Rutherford Diego de Las Casas Lisa Anne Hendricks Johannes Welbl Aidan Clark et al. 2022. Training compute-optimal large language models. arXiv:2203.15556. Retrieved from https:\/\/arxiv.org\/abs\/2203.15556","DOI":"10.52202\/068431-2176"},{"key":"e_1_3_1_24_2","unstructured":"Glenn Jocher Ayush Chaurasia and Jing Qiu. 2023. Ultralytics YOLOv8. GitHub Repository. Retrieved from https:\/\/github.com\/ultralytics\/ultralytics"},{"key":"e_1_3_1_25_2","unstructured":"Jared Kaplan Sam McCandlish Tom Henighan Tom B. Brown Benjamin Chess Rewon Child Scott Gray Alec Radford Jeffrey Wu and Dario Amodei. 2020. Scaling laws for neural language models. arXiv:2001.08361. Retrieved from https:\/\/arxiv.org\/abs\/2001.08361"},{"key":"e_1_3_1_26_2","volume-title":"Nonlinear Systems","author":"Khalil Hassan K.","year":"2002","unstructured":"Hassan K. Khalil. 2002. Nonlinear Systems (3rd ed.). Prentice-Hall, Upper Saddle River, NJ.","edition":"3"},{"key":"e_1_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.1145\/2228360.2228484"},{"key":"e_1_3_1_28_2","article-title":"BEVFormer: Learning bird\u2019s-eye-view representation from lidar-camera via spatiotemporal transformers","author":"Li Zhiqi","year":"2024","unstructured":"Zhiqi Li, Wenhai Wang, Hongyang Li, Enze Xie, Chonghao Sima, Tong Lu, Qiao Yu, and Jifeng Dai. 2024. BEVFormer: Learning bird\u2019s-eye-view representation from lidar-camera via spatiotemporal transformers. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2024.","journal-title":"IEEE Transactions on Pattern Analysis and Machine Intelligence"},{"key":"e_1_3_1_29_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.298"},{"key":"e_1_3_1_30_2","doi-asserted-by":"publisher","DOI":"10.1007\/s11554-023-01403-7"},{"key":"e_1_3_1_31_2","doi-asserted-by":"publisher","DOI":"10.1109\/TAC.2005.851439"},{"key":"e_1_3_1_32_2","volume-title":"Modern Control Engineering","author":"Ogata Katsuhiko","year":"2010","unstructured":"Katsuhiko Ogata. 2010. Modern Control Engineering (5th ed.). Prentice-Hall, Upper Saddle River, NJ.","edition":"5"},{"key":"e_1_3_1_33_2","doi-asserted-by":"publisher","DOI":"10.1109\/CSPA.2009.5069228"},{"key":"e_1_3_1_34_2","first-page":"77","volume-title":"NeurIPS 2019 Competition and Demonstration Track","author":"O\u2019Kelly Matthew","year":"2020","unstructured":"Matthew O\u2019Kelly, Hongrui Zheng, Dhruv Karthik, and Rahul Mangharam. 2020. F1TENTH: An open-source evaluation environment for continuous control and reinforcement learning. In NeurIPS 2019 Competition and Demonstration Track, 77\u201389. Retrieved from https:\/\/proceedings.mlr.press\/v123\/o-kelly20a.html"},{"key":"e_1_3_1_35_2","unstructured":"PyTorch Contributors. 2020. Pruning Tutorial. Retrieved April 7 2025 from https:\/\/pytorch.org\/tutorials\/intermediate\/pruning_tutorial.html"},{"key":"e_1_3_1_36_2","doi-asserted-by":"crossref","unstructured":"Joseph Redmon Santosh Kumar Divvala Ross B. Girshick and Ali Farhadi. 2015. You only look once: Unified real-time object detection. arXiv:1506.02640. Retrieved from http:\/\/arxiv.org\/abs\/1506.02640","DOI":"10.1109\/CVPR.2016.91"},{"key":"e_1_3_1_37_2","doi-asserted-by":"publisher","DOI":"10.1109\/RTAS.2016.7461344"},{"key":"e_1_3_1_38_2","doi-asserted-by":"publisher","DOI":"10.1109\/JPROC.2017.2779456"},{"key":"e_1_3_1_39_2","doi-asserted-by":"crossref","first-page":"54","DOI":"10.1016\/j.cogr.2023.04.001","article-title":"Artificial intelligence, machine learning and deep learning in advanced robotics, a review","volume":"3","author":"Soori Mohsen","year":"2023","unstructured":"Mohsen Soori, Behrooz Arezoo, and Roza Dastres. 2023. Artificial intelligence, machine learning and deep learning in advanced robotics, a review. Cognitive Robotics 3 (2023), 54\u201370.","journal-title":"Cognitive Robotics"},{"key":"e_1_3_1_40_2","doi-asserted-by":"publisher","DOI":"10.1109\/3DV.2017.00012"},{"key":"e_1_3_1_41_2","unstructured":"Yangzhen Wu Zhiqing Sun Shanda Li Sean Welleck and Yiming Yang. 2024. Inference scaling laws: An empirical analysis of compute-optimal inference for problem-solving with language models. arXiv:2408.00724. Retrieved from https:\/\/arxiv.org\/abs\/2408.00724"},{"key":"e_1_3_1_42_2","doi-asserted-by":"publisher","DOI":"10.3390\/app12168239"},{"key":"e_1_3_1_43_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-45329-8_15"},{"key":"e_1_3_1_44_2","doi-asserted-by":"publisher","DOI":"10.1145\/3566097.3567848"},{"key":"e_1_3_1_45_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCPS61052.2024.00040"},{"key":"e_1_3_1_46_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICAA64256.2024.00018"},{"key":"e_1_3_1_47_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-025-88184-0"},{"key":"e_1_3_1_48_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICPADS60453.2023.00184"}],"container-title":["ACM Transactions on Cyber-Physical Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3828657","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,21]],"date-time":"2026-09-21T13:14:50Z","timestamp":1789996490000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3828657"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,9,21]]},"references-count":47,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2026,10,31]]}},"alternative-id":["10.1145\/3828657"],"URL":"https:\/\/doi.org\/10.1145\/3828657","relation":{},"ISSN":["2378-962X","2378-9638"],"issn-type":[{"value":"2378-962X","type":"print"},{"value":"2378-9638","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,9,21]]},"assertion":[{"value":"2025-07-27","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-05-11","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-09-21","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}