The DOI matches its Crossref record.
A 1.06-to-5.09 TOPS/W reconfigurable hybrid-neural-network processor for deep learning applications
Abstract
An energy-efficient hybrid neural network (NN) processor is implemented in a 65nm technology. It has two 16×16 reconfigurable heterogeneous processing elements (PEs)arrays. To accelerate a hybrid-NN, the PE array is designed to support on demand partitioning and reconfiguration for parallel processing different NNs. To improve energy efficiency, each PE supports bit-width adaptive computing to meet variant bit-width of different neural layers. Measurement results show that this processor achieves a peak 409.6GOPS running at 200MHz and at most 5.09TOPS/W energy efficiency. This processor outperforms the state-of-the-art up to 5.2X in energy efficiency.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Some studies proposed to implement the argmax() function via hardware128–148, which could be beneficial to reduce the total transistor count and power consumption while at the same time increasing the throughput.”
From Hardware implementation of memristor-based artificial neural networks (2024) · cites this paper as [130]
Citations
Cites
The publisher hasn’t deposited this paper’s reference list.
Research works printed but not linked: 4
- chen, 2016, ICCAD
- moons, 2016, VLSI Symp
- chen, 2016, ISSCC
- donahue, 2015, WACV
Cited by
Eyeriss v2: A Flexible Accelerator for Emerging Deep Neural Networks on Mobile Devices
Chen, Y.-H., Yang, T.-J., Emer, J. S., & Sze, V. (2019) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2019.2910232 · Cited by 1,106
Hardware implementation of memristor-based artificial neural networks
Aguirre, F., Sebastian, A., Le Gallo, M., Song, W., Wang, T., Yang, J. J., Lu, W., Chang, M.-F., Ielmini, D., Yang, Y., et al. (2024) · Nature Communications · doi:10.1038/s41467-024-45670-9 · Cited by 564
Efficient Acceleration of Deep Learning Inference on Resource-Constrained Edge Devices: A Review
Shuvo, M. M. H., Islam, S. K., Cheng, J., & Morshed, B. I. (2023) · Proceedings of the IEEE · doi:10.1109/jproc.2022.3226481 · Cited by 436
“In [321], a reconfigurable ASIC implemented in the 65-nm process is presented targeting hybrid NNs.”
MethodUNPU: An Energy-Efficient Deep Neural Network Accelerator With Fully Variable Weight Bit Precision
Lee, J., Kim, C., Kang, S., Shin, D., Kim, S., & Yoo, H.-J. (2019) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2018.2865489 · Cited by 335
Lee, J., Kim, C., Kang, S., Shin, D., Kim, S., & Yoo, H.-J. (2018) · 2018 IEEE International Solid - State Circuits Conference - (ISSCC) · doi:10.1109/isscc.2018.8310262 · Cited by 303
“[5] S.”
Aimar, A., Mostafa, H., Calabrese, E., Rios-Navarro, A., Tapiador-Morales, R., Lungu, I.-A., Milde, M. B., Corradi, F., Linares-Barranco, A., Liu, S.-C., et al. (2019) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2018.2852335 · Cited by 292
Capra, M., Bussolino, B., Marchisio, A., Masera, G., Martina, M., & Shafique, M. (2020) · IEEE Access · doi:10.1109/access.2020.3039858 · Cited by 222
“For example, in [135], there are two 16x16 arrays, whose PEs are divided into general PEs and super PEs.”
BackgroundYin, S., Ouyang, P., Tang, S., Tu, F., Li, X., Zheng, S., Lu, T., Gu, J., Liu, L., & Wei, S. (2018) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2017.2778281 · Cited by 212
“This paper presents a hybrid-NN processor (called “Thinker”) [20], which is designed to address the three special features of hybrid-NNs.”
Self-citationCapra, M., Bussolino, B., Marchisio, A., Shafique, M., Masera, G., & Martina, M. (2020) · Future Internet · doi:10.3390/fi12070113 · Cited by 195
“In [90], an ASIC reconfigurable processor targeting hybrid NNs, i.e., networks with different layers, is presented.”
BackgroundEfficient Processing of Deep Neural Networks
Sze, V., Chen, Y., Yang, T.-J., & Emer, J. (2020) · Synthesis lectures on computer architecture · doi:10.2200/s01004ed1v01y202004cac050 · Cited by 183
Edge Intelligence: Concepts, Architectures, Applications, and Future Directions
Mendez, J., Bierzynski, K., Cuéllar, M. P., & Morales, D. P. (2022) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3486674 · Cited by 129
“In the work by Qiu et al. [151], another DL system based on FPGA is presented for CNN acceleration for ImageNet large-scale image classification.”
BackgroundMethodInfluentialUeyoshi, K., Ando, K., Hirose, K., Takamaeda-Yamazaki, S., Kadomoto, J., Miyata, T., Hamada, M., Kuroda, T., & Motomura, M. (2018) · Digest of technical papers/Digest of technical papers - IEEE International Solid-State Circuits Conference · doi:10.1109/isscc.2018.8310261 · Cited by 95
“These mechanisms are key enablers for handling various DNNs on a single homogeneous PE array (whereas [5] and [6] use hybrid cores and hybrid PEs, respectively, for different DNN types): e.”
BackgroundMethodEdge Intelligence: A Review of Deep Neural Network Inference in Resource-Limited Environments
Ngo, D., Park, H.-C., & Kang, B. (2025) · Electronics · doi:10.3390/electronics14122495 · Cited by 151
McDRAM: Low Latency and Energy-Efficient Matrix Computations in DRAM
Shin, H., Kim, D., Park, E., Park, S., Park, Y.-S., & Yoo, S. (2018) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2018.2857044 · Cited by 89
NoC-based DNN accelerator: a future design paradigm
Chen, K.-C. J., Ebrahimi, M., Wang, T., & Yang, Y.-C. (2019) · IEEE/ACM International Symposium on Networks-on-Chip (NOCS) · doi:10.1145/3313231.3352376 · Cited by 72
“In ASIC-based designs, to reach high performance and power efficiency, on-chip computing units are optimized for a particular application [20][21][22][23].”
BackgroundKadetotad, D. V., Yin, S., Berisha, V., Chakrabarti, C., & Seo, J.-S. (2020) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2020.2992900 · Cited by 68
“TableIshows a detailed comparison with prior ASIC and FPGA hardware designs for RNNs. Compared with the RNN ASIC works of [12] and [13], this work shows 2.90 × and 1.75 × higher energy efficiency (TOPS/W), respectively.”
MethodInfluentialSelf-citationEyeriss v2: A Flexible and High-Performance Accelerator for Emerging Deep Neural NetworksPreprint
Chen, Y., Emer, J., & Sze, V. (2018) · arXiv (Cornell University) · Cited by 63
Genc, H., Haj-Ali, A., Iyer, V., Amid, A., Mao, H., Wright, J., Schmidt, C., Zhao, J., Ou, A. J., Banister, M., et al. (2019) · arXiv (Cornell University) · Cited by 68
Ueyoshi, K., Ando, K., Hirose, K., Takamaeda-Yamazaki, S., Hamada, M., Kuroda, T., & Motomura, M. (2018) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2018.2871623 · Cited by 55
“) DNN accelerators [17], [23]–[25], using LUT-based or linear quantization, with this paper.”
Efficient Softmax Hardware Architecture for Deep Neural Networks
Du, G., Tian, C., Li, Z., Zhang, D., Yin, Y., & Ouyang, Y. (2019) · ACM Great Lakes Symposium on VLSI (GLSVLSI) · doi:10.1145/3299874.3317988 · Cited by 55
Neuromorphic computing's yesterday, today, and tomorrow – an evolutional view
Chen, Y., Li, H. H., Wu, C., Song, C., Li, S. K., Min, C., Cheng, H.-P., Wen, W., & Liu, X. (2017) · Integration · doi:10.1016/j.vlsi.2017.11.001 · Cited by 51
A Scalable Multi-TeraOPS Core for AI Training and Inference
Shukla, S., Fleischer, B., Ziegler, M. M., Silberman, J. A., Oh, J., Srinivasan, V., Choi, J., Mueller, S. M., Agrawal, A., Babinsky, T., et al. (2018) · IEEE Solid-State Circuits Letters · doi:10.1109/lssc.2019.2902738 · Cited by 39
“with 5 bits of exponent and 10 bits of mantissa (1, 5, 10) or the brain float (Bfloat 16) format with 8 bits of exponent and 7 bits of mantissa (1, 8, 7).”
BackgroundMemory-Efficient CNN Accelerator Based on Interlayer Feature Map Compression
Shao, Z., Chen, X., Du, L., Chen, L., Du, Y., Zhuang, W., Wei, H., Xie, C., & Wang, Z. (2021) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2021.3120312 · Cited by 41
Exploiting Retraining-Based Mixed-Precision Quantization for Low-Cost DNN Accelerator Design
Kim, N., Shin, D., Choi, W., Kim, G., & Park, J. (2020) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2020.3008996 · Cited by 39
Stream Processing Dual-Track CGRA for Object Inference
Fan, X., Wu, D., Cao, W., Luk, W. W., & Wang, L. (2018) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2018.2797600 · Cited by 38
“neural network processing unit [24], deep neural architecture [25], and hybrid-neural-network processor [26], are more suitable and energy efficient.”
BackgroundSecond-order Stencil Descent for Interior-point Hyperelasticity
Lan, L., Li, M., Jiang, C., Wang, H., & Yang, Y. (2023) · ACM Transactions on Graphics · doi:10.1145/3592104 · Cited by 47
“The potential of using the emerging neural processing unit (NPU) [Yin et al. 2017] for simulation is a worthy future topic.”
Method1.2 Intelligence on Silicon: From Deep-Neural-Network Accelerators to Brain Mimicking AI-SoCs
Yoo, H. (2019) · Digest of technical papers/Digest of technical papers - IEEE International Solid-State Circuits Conference · doi:10.1109/isscc.2019.8662469 · Cited by 39
“architecture in coarse grain with fine-grain bit precision [20].”
BackgroundA 2.17-mW Acoustic DSP Processor With CNN-FFT Accelerators for Intelligent Hearing Assistive Devices
Lee, Y.-C., Chi, T.-S., & Yang, C. (2020) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2020.2987695 · Cited by 38
“Table III shows the performance of the CNN processors [20], [22], [24], [37].”
BackgroundOttavi, G., Garofalo, A., Tagliavini, G., Conti, F., Mauro, A. D., Benini, L., & Rossi, D. (2023) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2023.3254810 · Cited by 41
“It employs a circuit-level voltage and frequency scaling technique to improve efficiency, capable of peak 76 GOPS and averages 2 TOPS/W. Thinker [20] employs configurable 2D arrays that can be partitioned into sub-arrays to compute different types of layers.”
MethodArchitecture of neural processing unit for deep neural networks
Lee, K. J. (2020) · Advances in computers · doi:10.1016/bs.adcom.2020.11.001 · Cited by 30
From Algorithm to Hardware: A Survey on Efficient and Safe Deployment of Deep Neural Networks
Geng, X., Wang, Z., Chen, C., Xu, Q., Xu, K., Jin, C., Gupta, M., Yang, X., Chen, Z., Sabry Aly, M. M., et al. (2025) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3394494 · Cited by 26
“In [202], 8-and 16-bit fixed-point NN layers were supported.”
Self-citationA Resources-Efficient Configurable Accelerator for Deep Convolutional Neural Networks
Hu, X., Zeng, Y., Li, Z., Zheng, X. W., Cai, S., & Xiong, X. (2019) · IEEE Access · doi:10.1109/access.2019.2919527 · Cited by 29
“Both designs in [10] and [24] are implemented with the TSMC 65-nm technology and have energy efficiency of 406.2 G-ops/s/W and 2.27 T-ops/s/W, respectively.”
BackgroundMethodInfluentialYang, J., Mao, R., Jiang, M., Cheng, Y., Sun, P.-S. V., Dong, S., Pedretti, G., Sheng, X., Ignowski, J., Li, H., et al. (2025) · Nature Communications · doi:10.1038/s41467-025-56254-6 · Cited by 38
Stripe: Tensor Compilation via the Nested Polyhedral ModelPreprint
Zerrell, T., & Bruestle, J. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1903.06498 · Cited by 48
“The industry is producing an explosion of varied and creative hardware accelerator architectures [6, 9, 14, 15, 22, 24, 36].”
BackgroundARA: Cross-Layer approximate computing framework based reconfigurable architecture for CNNs
Gong, Y., Liu, B., Ge, W., & Shi, L. (2019) · Microelectronics Journal · doi:10.1016/j.mejo.2019.03.011 · Cited by 21
The Development of Silicon for AI: Different Design Approaches
Lee, K. J., Lee, J., Choi, S., & Yoo, H. (2020) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2020.2996625 · Cited by 22
Chen, C., Ding, H., Peng, H., Zhu, H., Wang, Y., & Shi, C. ‐. R. (2018) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2018.2852780 · Cited by 21
PICACHU: Plug-In CGRA Handling Upcoming Nonlinear Operations in LLMs
Qin, J., Xia, T., Tan, C., Zhang, J. J., & Zhang, S. Q. (2025) · International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) · doi:10.1145/3676641.3716013 · Cited by 38
A Real-time Online Aircraft Neural Network System
Zhang, Y., Zhao, Q., Tao, L., Cao, J., Wei, M., & Zhang, X. (2019) · doi:10.1109/iwofc48002.2019.9078469 · Cited by 16
“PE, which is configured with a hybrid neural network processor to realize neural network acceleration[14].”
BackgroundMMNNN: A tree-based Multicast Mechanism for NoC-based deep Neural Network accelerators
Ouyang, Y., Tang, F., Chunlei, H., Zhou, W., & Wang, Q. (2021) · Microprocessors and Microsystems · doi:10.1016/j.micpro.2021.104242 · Cited by 15
Yang, E.-Y., Jia, T., Brooks, D. M., & Wei, G.-Y. (2021) · IEEE International Conference on Application-specific Systems, Architectures and Processors (ASAP) · doi:10.1109/asap52443.2021.00046 · Cited by 13
Linares-Barranco, A., Ríos-Navarro, A., Canas-Moreno, S., Piñero-Fuentes, E., Tapiador-Morales, R., & Delbrück, T. (2021) · International Conference on Neuromorphic Systems (ICONS) · doi:10.1145/3477145.3477167 · Cited by 13
Kawamoto, R., Taichi, M., Kabuto, M., Watanabe, D., Izumi, S., Yoshimoto, M., Kawaguchi, H., Matsukawa, G., Goto, T., & Kojima, M. (2020) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2020.2966331 · Cited by 13
Architecture, challenges and applications of dynamic reconfigurable computing
Lu, Y., Liu, L., Zhu, J., Yin, S., & Wei, S. (2020) · Journal of Semiconductors · doi:10.1088/1674-4926/41/2/021401 · Cited by 12
Self-citationKadetotad, D. V., Berisha, V., Chakrabarti, C., & Seo, J.-S. (2019) · IEEE Solid-State Circuits Letters · doi:10.1109/lssc.2019.2936761 · Cited by 12
“While recent ASIC designs targeting RNNs focus on improved energy-efficiency [6], [7], they do not incorporate compression techniques and do not report RNN accuracy for representative benchmarks, which are both necessary to accomplish practical ASR on small-form-factor edge devices.”
BackgroundMethodInfluentialAn Efficient Reconfigurable Framework for General Purpose CNN-RNN Models on FPGAs
Zeng, S., Guo, K., Fang, S., Kang, J., Xie, D., Shan, Y., Wang, Y., & Yang, H. (2018) · IEEE International Conference on Digital Signal Processing (DSP) · doi:10.1109/icdsp.2018.8631880 · Cited by 12
“TABLE IV: Comparison with Related Work This work VLSI’17[9] FPL’17[11] Hardware FPGA-ZU5EG ASIC FPGA-VC709 Frequency 200MHz 200MHz 100MHz Model LRCN LRCN LRCN Throughput 690.”
BackgroundMethodKadetotad, D. V., Berisha, V., Chakrabarti, C., & Seo, J.-S. (2019) · European Solid-State Circuits Conference (ESSCIRC) · doi:10.1109/esscirc.2019.8902809 · Cited by 11
Inference with Artificial Neural Networks on Analog Neuromorphic Hardware
Weis, J., Spilger, P., Billaudelle, S., Stradmann, Y., Emmel, A., Müller, E., Breitwieser, O., Grübl, A., Ilmberger, J., Karasenko, V., et al. (2020) · Communications in computer and information science · doi:10.1007/978-3-030-66770-2_15 · Cited by 15
Wiedemann, S., Shivapakash, S., Becking, D., Wiedemann, P., Samek, W., Gerfers, F., & Wiegand, T. (2021) · IEEE Open Journal of Circuits and Systems · doi:10.1109/ojcas.2021.3083332 · Cited by 11
“For 65nm comparison, in we have a gain of 3.1× when compared to EIE, 2.2× when compared to Eyeriss v2 and 2.4× when compared to Thinker.”
InfluentialAn Efficient CNN Inference Accelerator Based on Intra- and Inter-Channel Feature Map Compression
Xie, C., Shao, Z., Zhao, N. B., Du, Y., & Du, L. (2023) · IEEE Transactions on Circuits and Systems I Regular Papers · doi:10.1109/tcsi.2023.3287602 · Cited by 11
A 0.76 mm2 0.22 nJ/Pixel DL-Assisted 4K Video Encoder LSI for Quality-of-Experience Over Smartphones
Liu, T.-M., Tsai, C.-H., Wu, T.-H., Lin, J.-Y., Chen, L., Chou, H.-L., Chang, Y.-C., & Ju, C.-C. (2018) · IEEE Solid-State Circuits Letters · doi:10.1109/lssc.2019.2905958 · Cited by 9
Linares-Barranco, A., Ríos-Navarro, A., Tapiador-Morales, R., & Delbrück, T. (2019) · LA Referencia (Red Federada de Repositorios Institucionales de Publicaciones Científicas) · doi:10.5167/uzh-185140 · Cited by 13
Software-Defined Imaging: A Survey
Jayasuriya, S., Iqbal, O., Kodukula, V., Torres, V., LiKamWa, R., & Spanias, A. (2023) · Proceedings of the IEEE · doi:10.1109/jproc.2023.3266736 · Cited by 10
“Thinker [156] uses bit-width adaptive computing, on-demand array partitioning, and memory-gathering techniques to substantially reduce redundant memory accesses.”
BackgroundLow power & mobile hardware accelerators for deep convolutional neural networks
Scanlan, A. (2018) · Integration · doi:10.1016/j.vlsi.2018.11.010 · Cited by 11
A 2.17mW Acoustic DSP Processor with CNN-FFT Accelerators for Intelligent Hearing Aided Devices
Lee, Y., Chi, T.-S., & Yang, C. (2019) · IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas.2019.8771631 · Cited by 8
“ISSCC’16 [10] VLSI’16 [12] ISSCC’17 [11] VLSI’17 [13] This Work Technology (nm) 65 40 40 65 40 Core VDD (V) 0.”
BackgroundInfluentialTPE: A High-Performance Edge-Device Inference with Multi-level Transformational Mechanism
Wang, Z., Wei, J., Tang, X., Han, B., He, H., Liu, L., Wei, S., & Yin, S. (2023) · IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas57966.2023.10168614 · Cited by 7
“VLSI’17 [10] VLSI’17 [11] ISSCC’18 [12] This Work”
Self-citationRTPE: A High Energy Efficiency Inference Processor with RISC-V based Transformation Mechanism
Wang, Z., Du, H., Han, B., Xu, Y., Tang, X., Zhou, Y., Zheng, Z., Cui, W., Xiong, Y., Wei, S., et al. (2024) · IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas59952.2024.10595923 · Cited by 7
Self-citationDNNZip: Selective Layers Compression Technique in Deep Neural Network Accelerators
Lahdhiri, H., Palesi, M., Monteleone, S., Patti, D., Ascia, G., Lorandel, J., Bourdel, E., & Catania, V. (2020) · Euromicro Conference on Digital System Design (DSD) · doi:10.1109/dsd51259.2020.00088 · Cited by 7
Chung, S., & Wang, J. (2019) · IEEE Journal on Emerging and Selected Topics in Circuits and Systems · doi:10.1109/jetcas.2019.2934929 · Cited by 6
A CycleGAN Accelerator for Unsupervised Learning on Mobile Devices
Hsieh, Y.-Y., Lee, Y., & Yang, C. (2020) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas45731.2020.9180845 · Cited by 6
Wang, Z., Zhou, L., Xie, W., Chen, W., Su, J., Chen, W., Du, A., Li, S., Liang, M., Lin, Y.-J., et al. (2020) · Journal of Semiconductors · doi:10.1088/1674-4926/41/2/022401 · Cited by 6
On-chip Memory Optimized CNN Accelerator with Efficient Partial-sum Accumulation
Xu, H., Shiomi, J., & Onodera, H. (2020) · ACM Great Lakes Symposium on VLSI (GLSVLSI) · doi:10.1145/3386263.3406925 · Cited by 6
An Energy-Efficient Solid-State Organic Device Array for Neuromorphic Computing
Hu, L., Fattori, M., Schilp, W., Verbeek, R., Kazemzadeh, S., van de Burgt, Y., Kronemeijer, A. J., Gelinck, G. H., & Cantatore, E. (2023) · IEEE Transactions on Electron Devices · doi:10.1109/ted.2023.3327947 · Cited by 6
“Furthermore, inorganic technologies such as memristors, PCM, and SRAM [15] cannot be processed at a low temperature, thus preventing the use of flexible substrates.”
BackgroundThe Principle and Progress of Dynamically Reconfigurable Computing Technologies
Wei, S., & Lu, Y. (2020) · Chinese Journal of Electronics · doi:10.1049/cje.2020.05.002 · Cited by 5
Self-citationTanji: a General-purpose Neural Network Accelerator with Unified Crossbar Architecture
Zhu, H., Wang, Y., & Shi, C. ‐. R. (2019) · IEEE Design and Test · doi:10.1109/mdat.2019.2952329 · Cited by 5
“There are only several published works focusing on designing such unified hardware [6]–[8].”
BackgroundCo-design of deep neural nets and neural net accelerators for embedded vision applications
Amid, A., Kwon, K., Gholami, A., Wu, B., Asanović, K., & Keutzer, K. (2019) · IBM Journal of Research and Development · doi:10.1147/jrd.2019.2942284 · Cited by 6
Reconfigurable computing: toward software defined chips
Zhu, J., Wei, S., Liu, L., & Li, Z. (2020) · Scientia Sinica Informationis · doi:10.1360/ssi-2020-0130 · Cited by 4
Self-citationAn Arbitrary Kernel-size Applicable NoC-based DNN Processor Design with Hybrid Data Reuse
Chen, K.-C. J., Yang, Y.-C., & Liao, Y.-S. (2021) · IEEE International Midwest Symposium on Circuits and Systems (MWSCAS) · doi:10.1109/mwscas47672.2021.9531720 · Cited by 4
The position-based compression techniques for DNN model
Tang, M., Russo, E., & Palesi, M. (2023) · The Journal of Supercomputing · doi:10.1007/s11227-023-05339-4 · Cited by 3
“[3, 4, 8, 4, 8, 1, 5, 2, 2, 9, 3, 8, 14, 9, 3, 8, 2, 15, 9, 31, 1, 11, 5, 5, 2, 3, 1, 14, 7, 1, 4, 4, 8, 5, 1, 4, 2, 4, 2, 6, 2, 2, 8, 1, 7, 2, 10, 5, 2, 7, 2, 1, 4, 6, 2, 8, 1, 6, 1, 8, 2, 10, 3, 2, 3, 1, 7, 1, 7, 3, 6, 4, 4, 5, 3, 5, 3, 5, 1, 2, 1, 8, 6, 1, 8, 3, 5, 5, 2, 12, 12, 29, 33] Figure 6 shows the occurrenc…”
BackgroundA Scalable All-Digital Near-Memory Computing Architecture for Edge AIoT Applications
Nouripayam, M., Prieto, A., & Rodrigues, J. N. (2025) · IEEE Access · doi:10.1109/access.2025.3582013 · Cited by 3
Yang, J., Luo, X., Ke, Y., Wang, Z., Shang, H., Dong, S., Fu, Z., Yang, X., Liu, H., & Basu, A. (2025) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2025.3638445 · Cited by 5
In-memory Processing based on Time-domain Circuit
Kong, Y., & Yang, J. (2019) · ACM Great Lakes Symposium on VLSI (GLSVLSI) · doi:10.1145/3299874.3319484 · Cited by 2
“[2] proposed a reconfigurable heterogeneous DNN accelerator.”
Yang, C., Hou, J., Wang, Y., Zhang, H., Wang, X., & Geng, L. (2022) · Journal of Circuits Systems and Computers · doi:10.1142/s0218126622502899 · Cited by 2
Morphling: A Reconfigurable Architecture for Tensor Computation
Lu, L., & Liang, Y. E. (2021) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2021.3135322 · Cited by 2
“Recently, the demand for massive parallel computation has grown continuously in the field of CGRAs [16], [66], [70], [77], [78], [87].”
BackgroundWang, Z., Du, H., Xu, Y., Zhou, S., Zhou, J., Guo, L., Han, B., Tang, X., Qiao, S., Yin, S., et al. (2025) · IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS) · doi:10.1109/aicas64808.2025.11173136 · Cited by 1
Self-citationFragmented Huffman-Based Compression Methodology for CNN Targeting Resource-Constrained Edge Devices
Pal, C., Pankaj, S., Akram, W., Biswas, D., Mattela, G., & Acharyya, A. (2022) · Circuits Systems and Signal Processing · doi:10.1007/s00034-022-01968-x · Cited by 1
“Reference [5] also demonstrated the reduced precision support in recent hardware designs like Envision [22], Thinker [36], UNPU [19], Loom [28], and Stripes [17].”
MethodLLIE: Low-Light Image Enhancement Network Based on FPGA
Cui, J., Chen, G., Zhang, R., & Zhang, S. (2025) · Journal of Circuits Systems and Computers · doi:10.1142/s0218126625503505 · Cited by 1
Research on low-power neural network computing accelerator
Yan, J., Zhang, Y., Tu, F., Yang, J., Zheng, S., Ouyang, P., Liu, L., Xie, Y., Wei, S., & Yin, S. (2019) · Scientia Sinica Informationis · doi:10.1360/n112018-00282 · Cited by 1
Self-citationYang, C., Hou, J., Wang, Y., Zhou, Q. H., & Geng, L. (2019) · IEEE International Conference on Electronics, Circuits and Systems (ICECS) · doi:10.1109/icecs46596.2019.8964688 · Cited by 1
“Therefore, CGRA is widely used in multiple systems to deal with increasingly complicated applications, such as Thinker [2] for machine learning.”
MethodA 0.7mm2 8.54mW FocusNet Display LSI for Power Reduction on OLED Smart-phones
Liu, T.-M., Tsai, C.-H., Shih, S., Chang, C.-K., Lin, J.-Y., Hsieh, W., Chang, Y.-C., & Ju, C.-C. (2019) · IEEE Asian Solid-State Circuits Conference (A-SSCC) · doi:10.1109/a-sscc47793.2019.9056910 · Cited by 1
MOSDA: On-Chip Memory Optimized Sparse Deep Neural Network Accelerator With Efficient Index Matching
Xu, H., Shiomi, J., & Onodera, H. (2020) · IEEE Open Journal of Circuits and Systems · doi:10.1109/ojcas.2020.3035402 · Cited by 2
Reconfigurable Neural Network Accelerator and Simulator for Model Implementation
Nakahara, Y., Kiyama, M., Amagasaki, M., Zhao, Q., & Iida, M. (2021) · IEICE Transactions on Fundamentals of Electronics Communications and Computer Sciences · doi:10.1587/transfun.2021vlp0012 · Cited by 1
Tsurumi, K., Suzuki, K., & Takeuchi, K. (2018) · IEEE Asian Solid-State Circuits Conference (A-SSCC) · doi:10.1109/asscc.2018.8579333
An Inference Accelerator Design for Sparse Convolution Neural Network
Wang, D., Wu, T., Tan, H., & Li, H. (2021) · International Conference on Computer Science, Electronic Information Engineering and Intelligent Control Technology (CEI) · doi:10.1109/cei52496.2021.9574454
Kuroda, T., & Yip, W.-Y. (2021) · Cambridge University Press eBooks · doi:10.1017/9781108893299.005
A Bottom-Up Methodology for the Fast Assessment of CNN Mappings on Energy-Efficient Accelerators
Devic, G., Sassatelli, G., & Gamatié, A. (2023) · Journal of Low Power Electronics and Applications · doi:10.3390/jlpea13010005
Chen, L.-C., Li, Z., Lin, Y.-J., Lee, K.-P., & Tang, K. (2022) · IEEE Asia Pacific Conference on Circuits and Systems (APCCAS) · doi:10.1109/apccas55924.2022.10090334
“JSSC’17 Eyeriss [1] JSSC’17 DSIP [5] VLSIC’17 [6] TVLSI’20 [7] TCAS-I’21 [8] This work (post-sim)”
InfluentialDesign of a Highly Efficient Reconfigurable Processing Unit for Neural Network Acceleration
Chang, Y.-N. (2023) · International Conference on Intelligent Systems and Image Processing (ICISIP) · doi:10.12792/icisip2023.029
Heterogeneous Multi-Chiplets Neural Network Accelerator
Guoyi, Z., Ma, S., Zhang, C., & Wang, B. (2023) · Journal of Computer-Aided Design & Computer Graphics · doi:10.3724/sp.j.1089.2023.19451
“为了实现神经网络的高效计算 , Yin 等 [8] 提出 技术 [10] 以及 AMD 公司的 Infinity Fabric 技 术等 .”
Étude d'architectures dédiées aux systèmes embarqués intelligents et efficaces en énergie
Devic, G. (2021) · theses.fr (ABES) · doi:10.70675/afb9e868z24b8z4b45z8207z141c5dd1078d · Cited by 1
To Infinity and Beyond: A Reconfigurable Architecture for Space Vision Computing
Salvo, C. D., & Conficconi, D. (2026) · IEEE International Symposium on Field-Programmable Custom Computing Machines (FCCM) · doi:10.1109/fccm68464.2026.00081
“CGRAs bridge this gap by offering microsecond-scale re-configuration [7], but existing designs target neural-network inference [1] or video decoding [6].”
Sarma, B., Kumari, A., & Paily, R. P. (2026) · IEEE Guwahati Subsection Conference (GCON) · doi:10.1109/gcon69192.2026.11648302
“While architectures like Thinker [12] and Kraken [15] achieve massive peak throughputs, they do so by targeting deep, computationally exhaustive models (e.g., ResNet-50) using highly complex routing networks or massive spatial arrays.”
Metadata
- Authors
- Yin, S., Ouyang, P., Tang, S., Tu, F., Li, X., Liu, L., & Wei, S.
- Affiliations
- Tsinghua University (CN)
- Venue
- 2017 Symposium on VLSI Circuits
- Publisher
- IEEE
- Published
- 2017-06
- Pages
- C26-C27
- ISSN
- 2166-9600 (ISSN-L)
- OpenAlex
- W2744228767
- Cited by
- 93
- References
- 0
- Citations by year
- 2026: 2 · 2025: 7 · 2024: 3 · 2023: 10 · 2022: 6 · 2021: 11 · 2020: 18 · 2019: 20 · 2018: 14 · 2017: 2
- FWCI
- 20.89
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- Symposium on VLSI Circuits (publishedVersion)
- Rare & Special e-Zone (The Hong Kong University of Science and Technology) (submittedVersion)
- Topics
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
- Physical Sciences › Computer Science › Hardware and Architecture › Embedded Systems Design Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- reconfigurable processing elements, energy efficiency, deep learning acceleration, 65 nm CMOS
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.