{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T14:20:29Z","timestamp":1773843629326,"version":"3.50.1"},"reference-count":43,"publisher":"Association for Computing Machinery (ACM)","issue":"5","funder":[{"DOI":"10.13039\/501100003052","name":"MOTIE","doi-asserted-by":"crossref","award":["1415181094"],"award-info":[{"award-number":["1415181094"]}],"id":[{"id":"10.13039\/501100003052","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100020088","name":"KSRC","doi-asserted-by":"crossref","award":["20019956"],"award-info":[{"award-number":["20019956"]}],"id":[{"id":"10.13039\/501100020088","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Yonsei University Research Fund","award":["2024-22-0045"],"award-info":[{"award-number":["2024-22-0045"]}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Des. Autom. Electron. Syst."],"published-print":{"date-parts":[[2025,9,30]]},"abstract":"<jats:p>In electronic design automation (EDA), simulation models are often non-differentiable, and many design choices are discrete. As a result, greedy optimization methods based on numerical gradients are widely used, although they often lead to suboptimal solutions. In contrast, analytical methods may provide better solutions but require significant research effort. Reinforcement learning (RL) has been employed to address this problem; however, RL also suffers from notorious sample inefficiency, which is exaggerated in EDA because data sampling in EDA is very expensive due to slow simulations. This article proposes an alternative to RL for EDA, namely analytic gradient descent (AGD). Our method starts with a differentiable performance model, which can be either a learned surrogate or a static model. It then applies transformations similar to Shannon decomposition for each design variable in the performance model. Finally, one design option for each variable is selected using a one-hot variable, which is trained via a straight-through estimator (STE) through gradient descent. We demonstrate AGD on the well-known gate sizing problem using both a learned surrogate and a static model across 20 industrial benchmark circuits. Our experimental results show that the proposed method can outperform a several-decade-old commercial tool in the gate sizing task for 19 out of the 20 circuits.<\/jats:p>\n          <jats:p\/>","DOI":"10.1145\/3748257","type":"journal-article","created":{"date-parts":[[2025,7,17]],"date-time":"2025-07-17T11:31:07Z","timestamp":1752751867000},"page":"1-22","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":1,"title":["AGD: Analytic Gradient Descent for Discrete Optimization in EDA and its Use to Gate Sizing"],"prefix":"10.1145","volume":"30","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6388-1883","authenticated-orcid":false,"given":"Phuoc","family":"Pham","sequence":"first","affiliation":[{"name":"Technical University of Munich","place":["Munich, Germany"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-8984-6351","authenticated-orcid":false,"given":"Tae-Min","family":"Park","sequence":"additional","affiliation":[{"name":"Yonsei University","place":["Seoul, Korea (the Republic of)"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-2647-9661","authenticated-orcid":false,"given":"Sung-Hyuk","family":"Cho","sequence":"additional","affiliation":[{"name":"Yonsei University","place":["Seoul, Korea (the Republic of)"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8853-305X","authenticated-orcid":false,"given":"Tayyeb","family":"Mahmood","sequence":"additional","affiliation":[{"name":"Nextwave Co.","place":["Daejeon, Korea (the Republic of)"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1502-5353","authenticated-orcid":false,"given":"Joon-Sung","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Electrical and Electronic Engineering, Department of Systems Semiconductor Engineering, BK21 Graduate Program in Intelligent Semiconductor Technology, Yonsei University","place":["Seoul, Korea (the Republic of)"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5819-1995","authenticated-orcid":false,"given":"Jaeyong","family":"Chung","sequence":"additional","affiliation":[{"name":"Department of Electrical and Electronic Engineering, Department of Systems Semiconductor Engineering, Yonsei University","place":["Seoul, Korea (the Republic of)"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,8,12]]},"reference":[{"key":"e_1_3_1_2_2","unstructured":"[n.d.]. cppyy: Automatic Python-C++ bindings. Retrieved April 07 2024 from https:\/\/cppyy.readthedocs.io\/en\/latest\/"},{"key":"e_1_3_1_3_2","unstructured":"[n.d.]. OpenROAD. Available at Retrieved November 18 2022 from https:\/\/github.com\/The-OpenROAD-Project\/OpenROAD"},{"key":"e_1_3_1_4_2","unstructured":"[n.d.]. Parallax Static Timing Analyzer. Retrieved November 18 2022 from https:\/\/github.com\/The-OpenROAD-Project\/OpenSTA (2022\/11\/18)."},{"key":"e_1_3_1_5_2","doi-asserted-by":"publisher","DOI":"10.1145\/3400302.3415690"},{"key":"e_1_3_1_6_2","volume-title":"Proceedings of the NeurIPS 2020 Workshop on Machine Learning for Systems","author":"Agnesina Anthony","year":"2020","unstructured":"Anthony Agnesina, Sai Pentapati, and Sung Kyu Lim. 2020. A general framework for VLSI tool parameter optimization with deep reinforcement learning. In Proceedings of the NeurIPS 2020 Workshop on Machine Learning for Systems."},{"key":"e_1_3_1_7_2","doi-asserted-by":"publisher","DOI":"10.1145\/3534678.3539220"},{"key":"e_1_3_1_8_2","doi-asserted-by":"publisher","DOI":"10.1145\/3316781.3317857"},{"key":"e_1_3_1_9_2","unstructured":"Yoshua Bengio Nicholas L\u00e9onard and Aaron Courville. 2013. Estimating or propagating gradients through stochastic neurons for conditional computation. arXiv preprint arXiv:1308.3432"},{"key":"e_1_3_1_10_2","doi-asserted-by":"publisher","DOI":"10.1109\/ASP-DAC52403.2022.9712578"},{"key":"e_1_3_1_11_2","doi-asserted-by":"publisher","DOI":"10.5555\/3540261.3541523"},{"key":"e_1_3_1_12_2","doi-asserted-by":"publisher","DOI":"10.1109\/43.365122"},{"key":"e_1_3_1_13_2","doi-asserted-by":"publisher","DOI":"10.5555\/3495724.3497383"},{"key":"e_1_3_1_14_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICICS52457.2021.9464626"},{"key":"e_1_3_1_15_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-022-05172-4"},{"key":"e_1_3_1_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/MLCAD48534.2019.9142109"},{"key":"e_1_3_1_17_2","doi-asserted-by":"publisher","DOI":"10.1145\/3489517.3530486"},{"key":"e_1_3_1_18_2","doi-asserted-by":"crossref","unstructured":"Zizheng Guo Mingjie Liu Jiaqi Gu Shuhan Zhang David Z. Pan and Yibo Lin. 2022. A timing engine inspired graph neural network model for pre-routing slack prediction. In Proceedings of the 59th ACM\/IEEE Design Automation Conference. 1207\u20131212.","DOI":"10.1145\/3489517.3530597"},{"key":"e_1_3_1_19_2","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS.2018.8351885"},{"key":"e_1_3_1_20_2","doi-asserted-by":"publisher","DOI":"10.1145\/3489517.3530598"},{"key":"e_1_3_1_21_2","doi-asserted-by":"publisher","DOI":"10.1109\/ASP-DAC47756.2020.9045559"},{"key":"e_1_3_1_22_2","unstructured":"Wenzel Jakob Jason Rhinelander and Dean Moldovan. 2016. pybind11 \u2014 Seamless operability between C++11 and Python. Retrieved April 19 2022 from https:\/\/github.com\/pybind\/pybind11"},{"key":"e_1_3_1_23_2","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-021-03819-2"},{"key":"e_1_3_1_24_2","doi-asserted-by":"publisher","DOI":"10.1145\/2160916.2160949"},{"key":"e_1_3_1_25_2","volume-title":"Proceedings of the 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020","author":"Kaiser Lukasz","year":"2020","unstructured":"Lukasz Kaiser, Mohammad Babaeizadeh, Piotr Milos, Blazej Osinski, Roy H. Campbell, Konrad Czechowski, Dumitru Erhan, Chelsea Finn, Piotr Kozakowski, Sergey Levine, Afroz Mohiuddin, Ryan Sepassi, George Tucker, and Henryk Michalewski. [2020]. Model based reinforcement learning for atari. In Proceedings of the 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020."},{"key":"e_1_3_1_26_2","doi-asserted-by":"publisher","DOI":"10.1007\/s10878-009-9267-0"},{"key":"e_1_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.1145\/123186.123301"},{"key":"e_1_3_1_28_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAD45719.2019.8942063"},{"key":"e_1_3_1_29_2","doi-asserted-by":"publisher","DOI":"10.1109\/DAC18074.2021.9586138"},{"key":"e_1_3_1_30_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCD.2004.1347933"},{"key":"e_1_3_1_31_2","unstructured":"Azalia Mirhoseini Anna Goldie Mustafa Yazgan Joe Jiang Ebrahim Songhori Shen Wang Young-Joon Lee Eric Johnson Omkar Pathak Sungmin Bae et\u00a0al. 2020. Chip placement with deep reinforcement learning. arXiv preprint arXiv:2004.10746 (2020)."},{"key":"e_1_3_1_32_2","first-page":"2430","volume-title":"Proceedings of the International Conference on Machine Learning","author":"Mirhoseini Azalia","year":"2017","unstructured":"Azalia Mirhoseini, Hieu Pham, Quoc V. Le, Benoit Steiner, Rasmus Larsen, Yuefeng Zhou, Naveen Kumar, Mohammad Norouzi, Samy Bengio, and Jeff Dean. 2017. Device placement optimization with reinforcement learning. In Proceedings of the International Conference on Machine Learning. PMLR, 2430\u20132439."},{"key":"e_1_3_1_33_2","doi-asserted-by":"publisher","DOI":"10.1145\/3489517.3530645"},{"key":"e_1_3_1_34_2","doi-asserted-by":"publisher","DOI":"10.1145\/3400302.3415710"},{"key":"e_1_3_1_35_2","unstructured":"Adam Paszke Sam Gross Francisco Massa Adam Lerer James Bradbury Gregory Chanan Trevor Killeen Zeming Lin Natalia Gimelshein Luca Antiga et\u00a0al. 2019. Pytorch: An imperative style high-performance deep learning library. Proceedings of the 33rd International Conference on Neural Information Processing Systems 32 (2019) 8026\u20138037."},{"key":"e_1_3_1_36_2","doi-asserted-by":"crossref","unstructured":"Satwik Patnaik Vasudev Gohil Hao Guo and Jeyavijayan J. V. Rajendran. 2022. Reinforcement learning for hardware security: Opportunities developments and challenges. In 2022 19th International SoC Design Conference (ISOCC). 217\u2013218.","DOI":"10.1109\/ISOCC56007.2022.10031569"},{"key":"e_1_3_1_37_2","doi-asserted-by":"publisher","DOI":"10.1109\/DAC56929.2023.10247924"},{"key":"e_1_3_1_38_2","doi-asserted-by":"publisher","DOI":"10.1038\/nature16961"},{"key":"e_1_3_1_39_2","doi-asserted-by":"publisher","DOI":"10.1109\/DAC18072.2020.9218757"},{"key":"e_1_3_1_40_2","unstructured":"Hanrui Wang Jiacheng Yang Hae-Seung Lee and Song Han. 2018. Learning to design circuits. arXiv preprint arXiv:1812.02734 (2018)."},{"key":"e_1_3_1_41_2","unstructured":"Minjie Wang Da Zheng Zihao Ye Quan Gan Mufei Li Xiang Song Jinjing Zhou Chao Ma Lingfan Yu Yu Gai et\u00a0al. 2019. Deep graph library: A graph-centric highly-performant package for graph neural networks. arXiv preprint arXiv:1909.01315 (2019)."},{"key":"e_1_3_1_42_2","doi-asserted-by":"publisher","DOI":"10.1145\/2593069.2593131"},{"key":"e_1_3_1_43_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2022.3149977"},{"key":"e_1_3_1_44_2","doi-asserted-by":"publisher","DOI":"10.1145\/3380446.3430622"}],"container-title":["ACM Transactions on Design Automation of Electronic Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3748257","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,12]],"date-time":"2025-08-12T13:16:24Z","timestamp":1755004584000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3748257"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,12]]},"references-count":43,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2025,9,30]]}},"alternative-id":["10.1145\/3748257"],"URL":"https:\/\/doi.org\/10.1145\/3748257","relation":{},"ISSN":["1084-4309","1557-7309"],"issn-type":[{"value":"1084-4309","type":"print"},{"value":"1557-7309","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8,12]]},"assertion":[{"value":"2025-01-13","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-06-24","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-08-12","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}