{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,11,11]],"date-time":"2025-11-11T14:04:24Z","timestamp":1762869864505,"version":"build-2065373602"},"reference-count":41,"publisher":"Association for Computing Machinery (ACM)","issue":"1","content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Des. Autom. Electron. Syst."],"published-print":{"date-parts":[[2026,1,31]]},"abstract":"<jats:p>\n                    ReRAM-based Processing-In-Memory (PIM) architectures are compelling contenders for deep learning due to their ability to perform matrix-vector multiplications (MVMs) directly within the memory, significantly reducing data movement and enhancing computational efficiency. However, since MVMs occur in the analog domain, analog-to-digital converters (ADCs) dominate power consumption and area overhead in current implementations. To this end, we propose RedPIM, an efficient ReRAM-based PIM accelerator design for deep neural networks (DNNs) that reduces the number of analog-to-digital conversions. RedPIM exploits the fact that in ReRAM-based PIM accelerators, the overall energy consumption generally increases with the number of activated analog-to-digital conversions. Specifically, we introduce a novel training algorithm that is aware of the ADC overhead during activation value quantization and optimizes accuracy concurrently. From a hardware design perspective, we develop a lookup table (LUT)-based quantization module to enable efficient and low-cost activation value quantization. In addition, we propose an efficient adaptive operation unit (OU) size assignment scheme that further minimizes analog-to-digital conversions by considering activation sparsity and weight distribution. Extensive experimental results show our RedPIM reduces latency to 27.72% and energy consumption to 10.15% of the baseline, with minimal accuracy loss, making it a promising solution for enhancing DNN acceleration. The code for this project is available at:\n                    <jats:ext-link xmlns:xlink=\"http:\/\/www.w3.org\/1999\/xlink\" xlink:href=\"https:\/\/github.com\/JialeLiLab\/ADC_aware_Learning.git\">https:\/\/github.com\/JialeLiLab\/ADC_aware_Learning.git<\/jats:ext-link>\n                    .\n                  <\/jats:p>","DOI":"10.1145\/3769122","type":"journal-article","created":{"date-parts":[[2025,9,20]],"date-time":"2025-09-20T11:00:33Z","timestamp":1758366033000},"page":"1-26","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["RedPIM: An Efficient PIM Accelerator Design with Reduced Analog-to-Digital Conversions"],"prefix":"10.1145","volume":"31","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-3698-7603","authenticated-orcid":false,"given":"Jiale","family":"Li","sequence":"first","affiliation":[{"name":"School of Computer Science, The University of Auckland","place":["Auckland, New Zealand"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-8471-2933","authenticated-orcid":false,"given":"Yulin","family":"Fu","sequence":"additional","affiliation":[{"name":"School of Computer Science, The University of Auckland","place":["Auckland, New Zealand"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0854-039X","authenticated-orcid":false,"given":"Sean Longyu","family":"Ma","sequence":"additional","affiliation":[{"name":"School of Computer Science, The University of Auckland","place":["Auckland, New Zealand"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7007-6746","authenticated-orcid":false,"given":"Chiu-Wing","family":"Sham","sequence":"additional","affiliation":[{"name":"School of Computer Science, The University of Auckland","place":["Auckland, New Zealand"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4549-744X","authenticated-orcid":false,"given":"Chong","family":"Fu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Northeastern University","place":["Shenyang, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,11,11]]},"reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","DOI":"10.1145\/3297858.3304049"},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAD51958.2021.9643502"},{"issue":"7","key":"e_1_3_2_4_2","doi-asserted-by":"crossref","first-page":"1414","DOI":"10.1109\/TCAD.2019.2917852","article-title":"Low bit-width convolutional neural network on RRAM","volume":"39","author":"Cai Yi","year":"2019","unstructured":"Yi Cai, Tianqi Tang, Lixue Xia, Boxun Li, Yu Wang, and Huazhong Yang. 2019. Low bit-width convolutional neural network on RRAM. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 39, 7 (2019), 1414\u20131427.","journal-title":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems"},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00242"},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2018.2789723"},{"key":"e_1_3_2_7_2","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001140"},{"key":"e_1_3_2_8_2","unstructured":"Jungwook Choi Zhuo Wang Swagath Venkataramani Pierce I-Jen Chuang Vijayalakshmi Srinivasan and Kailash Gopalakrishnan. 2018. Pact: Parameterized clipping activation for quantized neural networks. arXiv:1805.06085. Retrieved from https:\/\/arxiv.org\/abs\/1805.06085"},{"key":"e_1_3_2_9_2","doi-asserted-by":"publisher","DOI":"10.23919\/DATE51398.2021.9474181"},{"key":"e_1_3_2_10_2","first-page":"97","volume-title":"Proceedings of the 59th ACM\/IEEE Design Automation Conference","author":"He Yintao","year":"2022","unstructured":"Yintao He, Songyun Qu, Ying Wang, Bing Li, Huawei Li, and Xiaowei Li. 2022. InfoX: An energy-efficient ReRAM accelerator design with information-lossless low-bit ADCs. In Proceedings of the 59th ACM\/IEEE Design Automation Conference. 97\u2013102."},{"issue":"7","key":"e_1_3_2_11_2","first-page":"2115","article-title":"Saving energy of RRAM-based neural accelerator through state-aware computing","volume":"41","author":"He Yintao","year":"2021","unstructured":"Yintao He, Ying Wang, Huawei Li, and Xiaowei Li. 2021. Saving energy of RRAM-based neural accelerator through state-aware computing. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 41, 7 (2021), 2115\u20132127.","journal-title":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems"},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1145\/3500929"},{"key":"e_1_3_2_13_2","doi-asserted-by":"publisher","DOI":"10.1145\/3394885.3431554"},{"key":"e_1_3_2_14_2","unstructured":"Yulhwa Kim Hyungjun Kim and Jae-Joon Kim. 2018. Neural network-hardware co-design for scalable RRAM-based BNN accelerators. arXiv:1811.02187. Retrieved from https:\/\/arxiv.org\/abs\/1811.02187"},{"key":"e_1_3_2_15_2","doi-asserted-by":"publisher","DOI":"10.1145\/3528104"},{"key":"e_1_3_2_16_2","first-page":"856","volume-title":"Proceedings of the 2021 Design, Automation & Test in Europe Conference & Exhibition (DATE)","author":"Kim Yulhwa","year":"2021","unstructured":"Yulhwa Kim, Hyungjun Kim, Jihoon Park, Hyunmyung Oh, and Jae-Joon Kim. 2021. Mapping binary ResNets on computing-in-memory hardware with low-bit ADCs. In Proceedings of the 2021 Design, Automation & Test in Europe Conference & Exhibition (DATE). IEEE, 856\u2013861."},{"key":"e_1_3_2_17_2","first-page":"1","volume-title":"Proceedings of the 2024 Design, Automation & Test in Europe Conference & Exhibition (DATE)","author":"Li Wanqian","year":"2024","unstructured":"Wanqian Li, Xiaotian Sun, Xinyu Wang, Lei Wang, Yinhe Han, and Xiaoming Chen. 2024. PIMSYN: Synthesizing processing-in-memory CNN accelerators. In Proceedings of the 2024 Design, Automation & Test in Europe Conference & Exhibition (DATE). IEEE, 1\u20136."},{"issue":"6","key":"e_1_3_2_18_2","first-page":"1871","article-title":"An ADC-less RRAM-based computing-in-memory macro with binary CNN for efficient edge AI","volume":"70","author":"Li Yi","year":"2023","unstructured":"Yi Li, Jia Chen, Linfang Wang, Woyu Zhang, Zeyu Guo, Jun Wang, Yongkang Han, Zhi Li, Fei Wang, Chunmeng Dou, et\u00a0al. 2023. An ADC-less RRAM-based computing-in-memory macro with binary CNN for efficient edge AI. IEEE Transactions on Circuits and Systems II: Express Briefs 70, 6 (2023), 1871\u20131875.","journal-title":"IEEE Transactions on Circuits and Systems II: Express Briefs"},{"key":"e_1_3_2_19_2","doi-asserted-by":"publisher","DOI":"10.1145\/3240765.3240800"},{"key":"e_1_3_2_20_2","doi-asserted-by":"crossref","unstructured":"Xiyan Liu and Qiang Liu. 2024. HNMC: Hybrid near-memory computing circuit for neural network acceleration. IEEE Transactions on Circuits and Systems II: Express Briefs 71 11 (2024) 4763\u20134767.","DOI":"10.1109\/TCSII.2024.3403830"},{"key":"e_1_3_2_21_2","doi-asserted-by":"crossref","first-page":"659060","DOI":"10.3389\/frai.2021.659060","article-title":"NeuroSim simulator for compute-in-memory hardware accelerator: Validation and benchmark","volume":"4","author":"Lu Anni","year":"2021","unstructured":"Anni Lu, Xiaochen Peng, Wantong Li, Hongwu Jiang, and Shimeng Yu. 2021. NeuroSim simulator for compute-in-memory hardware accelerator: Validation and benchmark. Frontiers in Artificial Intelligence 4 (2021), 659060.","journal-title":"Frontiers in Artificial Intelligence"},{"key":"e_1_3_2_22_2","doi-asserted-by":"crossref","first-page":"280","DOI":"10.1109\/ASPDAC.2016.7428024","volume-title":"Proceedings of the 2016 21st Asia and South Pacific Design Automation Conference (ASP-DAC)","author":"Ni Leibin","year":"2016","unstructured":"Leibin Ni, Yuhao Wang, Hao Yu, Wei Yang, Chuliang Weng, and Junfeng Zhao. 2016. An energy-efficient matrix multiplication accelerator by distributed in-memory computing on binary RRAM crossbar. In Proceedings of the 2016 21st Asia and South Pacific Design Automation Conference (ASP-DAC). IEEE, 280\u2013285."},{"key":"e_1_3_2_23_2","doi-asserted-by":"publisher","DOI":"10.1145\/3140659.3080254"},{"key":"e_1_3_2_24_2","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2011.2107214"},{"key":"e_1_3_2_25_2","doi-asserted-by":"crossref","first-page":"1042","DOI":"10.1145\/3673038.3673044","volume-title":"Proceedings of the 53rd International Conference on Parallel Processing","author":"Abrebekoh Mohammad Sabri","year":"2024","unstructured":"Mohammad Sabri Abrebekoh, Marc Riera Villanueva, and Antonio Gonzalez. 2024. ReDy: A novel ReRAM-centric dynamic quantization approach for energy-efficient CNNs. In Proceedings of the 53rd International Conference on Parallel Processing. 1042\u20131051."},{"key":"e_1_3_2_26_2","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001139"},{"key":"e_1_3_2_27_2","volume-title":"Image and Video Compression for Multimedia Engineering: Fundamentals, Algorithms, and Standards","author":"Shi Yun-Qing","year":"2019","unstructured":"Yun-Qing Shi and Huifang Sun. 2019. Image and Video Compression for Multimedia Engineering: Fundamentals, Algorithms, and Standards. CRC Press."},{"key":"e_1_3_2_28_2","first-page":"325","volume-title":"Proceedings of the 2020 25th Asia and South Pacific Design Automation Conference (ASP-DAC)","author":"Sun Hanbo","year":"2020","unstructured":"Hanbo Sun, Zhenhua Zhu, Yi Cai, Xiaoming Chen, Yu Wang, and Huazhong Yang. 2020. An energy-efficient quantized and regularized training framework for processing-in-memory accelerators. In Proceedings of the 2020 25th Asia and South Pacific Design Automation Conference (ASP-DAC). IEEE, 325\u2013330."},{"key":"e_1_3_2_29_2","doi-asserted-by":"publisher","DOI":"10.1109\/NVMSA.2017.8064475"},{"issue":"6","key":"e_1_3_2_30_2","first-page":"2249","article-title":"Low-power SAR ADC design: Overview and survey of state-of-the-art techniques","volume":"69","author":"Tang Xiyuan","year":"2022","unstructured":"Xiyuan Tang, Jiaxin Liu, Yi Shen, Shaolan Li, Linxiao Shen, Arindam Sanyal, Kareem Ragab, and Nan Sun. 2022. Low-power SAR ADC design: Overview and survey of state-of-the-art techniques. IEEE Transactions on Circuits and Systems I: Regular Papers 69, 6 (2022), 2249\u20132262.","journal-title":"IEEE Transactions on Circuits and Systems I: Regular Papers"},{"key":"e_1_3_2_31_2","first-page":"580","volume-title":"Proceedings of the 2024 IEEE International Solid-State Circuits Conference (ISSCC)","volume":"67","author":"Wen Tai-Hao","year":"2024","unstructured":"Tai-Hao Wen, Hung-Hsi Hsu, Win-San Khwa, Wei-Hsing Huang, Zhao-En Ke, Yu-Hsiang Chin, Hua-Jin Wen, Yu-Chen Chang, Wei-Ting Hsu, Chung-Chuan Lo, et\u00a0al. 2024. 34.8 A 22nm 16Mb floating-point ReRAM compute-in-memory macro with 31.2 TFLOPS\/W for AI edge devices. In Proceedings of the 2024 IEEE International Solid-State Circuits Conference (ISSCC), Vol. 67. IEEE, 580\u2013582."},{"key":"e_1_3_2_32_2","doi-asserted-by":"publisher","DOI":"10.1145\/2897937.2898101"},{"key":"e_1_3_2_33_2","doi-asserted-by":"crossref","unstructured":"Jiahong Xu Haikun Liu Zhuohui Duan Xiaofei Liao Hai Jin Xiaokang Yang Huize Li Cong Liu Fubing Mao and Yu Zhang. 2024. ReHarvest: An ADC resource-harvesting crossbar architecture for ReRAM-based DNN accelerators. ACM Transactions on Architecture and Code Optimization 21 3 (2024) 1\u201326.","DOI":"10.1145\/3659208"},{"key":"e_1_3_2_34_2","doi-asserted-by":"publisher","DOI":"10.1145\/3701034"},{"issue":"6","key":"e_1_3_2_35_2","first-page":"1916","article-title":"Multiplication through a single look-up-table (LUT) in CNN inference computation","volume":"41","author":"Xu Shiyu","year":"2021","unstructured":"Shiyu Xu, Qi Wang, Xingbo Wang, Shihang Wang, and Terry Tao Ye. 2021. Multiplication through a single look-up-table (LUT) in CNN inference computation. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 41, 6 (2021), 1916\u20131928.","journal-title":"IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems"},{"key":"e_1_3_2_36_2","doi-asserted-by":"publisher","DOI":"10.1145\/3307650.3322271"},{"key":"e_1_3_2_37_2","first-page":"1","volume-title":"Proceedings of the 2023 IEEE International Solid-State Circuits Conference (ISSCC)","author":"Yonar Abdullah Serdar","year":"2023","unstructured":"Abdullah Serdar Yonar, Pier Andrea Francese, Matthias Br\u00e4ndli, Marcel Kossel, Mridula Prathapan, Thomas Morf, Andrea Ruffino, and Taekwang Jang. 2023. An 8b 1.0-to-1.25 GS\/s 0.7-to-0.8 V single-stage time-based gated-ring-oscillator ADC with 2\u00d7 interpolating sense-amplifier-latches. In Proceedings of the 2023 IEEE International Solid-State Circuits Conference (ISSCC). IEEE, 1\u20133."},{"issue":"11","key":"e_1_3_2_38_2","doi-asserted-by":"crossref","first-page":"3466","DOI":"10.1109\/JSSC.2022.3162602","article-title":"A 65-nm 8T SRAM compute-in-memory macro with column ADCs for processing neural networks","volume":"57","author":"Yu Chengshuo","year":"2022","unstructured":"Chengshuo Yu, Taegeun Yoo, Kevin Tshun Chuan Chai, Tony Tae-Hyoung Kim, and Bongjin Kim. 2022. A 65-nm 8T SRAM compute-in-memory macro with column ADCs for processing neural networks. IEEE Journal of Solid-State Circuits 57, 11 (2022), 3466\u20133476.","journal-title":"IEEE Journal of Solid-State Circuits"},{"key":"e_1_3_2_39_2","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00029"},{"key":"e_1_3_2_40_2","doi-asserted-by":"crossref","unstructured":"Zihan Zhang Jianfei Jiang Qin Wang Zhigang Mao and Naifeng Jing. 2023. 3A-ReRAM: Adaptive activation accumulation in reram-based CNN accelerator. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems 43 1 (2023) 176\u2013188.","DOI":"10.1109\/TCAD.2023.3297968"},{"key":"e_1_3_2_41_2","volume-title":"Proceedings of the NeurIPS 2024 Workshop Machine Learning with New Compute Paradigms","author":"Zhao Lei","year":"2024","unstructured":"Lei Zhao, Luca Buonanno, Aishwarya Natarajan, Jim Ignowski, and Giacomo Pedretti. 2024. Noise aware finetuning for analog non-linear dot product engine. In Proceedings of the NeurIPS 2024 Workshop Machine Learning with New Compute Paradigms."},{"key":"e_1_3_2_42_2","first-page":"1","volume-title":"Proceedings of the 2020 57th ACM\/IEEE Design Automation Conference (DAC)","author":"Zheng Qilin","year":"2020","unstructured":"Qilin Zheng, Zongwei Wang, Zishun Feng, Bonan Yan, Yimao Cai, Ru Huang, Yiran Chen, Chia-Lin Yang, and Hai Helen Li. 2020. Lattice: An ADC\/DAC-less ReRAM-based processing-in-memory architecture for accelerating deep convolution neural networks. In Proceedings of the 2020 57th ACM\/IEEE Design Automation Conference (DAC). IEEE, 1\u20136."}],"container-title":["ACM Transactions on Design Automation of Electronic Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3769122","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,11]],"date-time":"2025-11-11T13:51:28Z","timestamp":1762869088000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3769122"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,11,11]]},"references-count":41,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2026,1,31]]}},"alternative-id":["10.1145\/3769122"],"URL":"https:\/\/doi.org\/10.1145\/3769122","relation":{},"ISSN":["1084-4309","1557-7309"],"issn-type":[{"type":"print","value":"1084-4309"},{"type":"electronic","value":"1557-7309"}],"subject":[],"published":{"date-parts":[[2025,11,11]]},"assertion":[{"value":"2025-05-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-09-15","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-11-11","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}