YodaNN: An Architecture for Ultralow Power Binary-Weight CNN Acceleration
Open accessThe DOI matches its Crossref record; 51 of the 51 references it cites were found in a registry or an index.
Abstract
Convolutional neural networks (CNNs) have revolutionized the world of computer vision over the last few years, pushing image classification beyond human accuracy. The computational effort of today's CNNs requires power-hungry parallel processors or GP-GPUs. Recent developments in CNN accelerators for system-on-chip integration have reduced energy consumption significantly. Unfortunately, even these highly optimized devices are above the power envelope imposed by mobile and deeply embedded applications and face hard limitations caused by CNN weight I/O and storage. This prevents the adoption of CNNs in future ultralow power Internet of Things end-nodes for near-sensor analytics. Recent algorithmic and theoretical advancements enable competitive classification accuracy even when limiting CNNs to binary (+1/-1) weights during training. These new findings bring major optimization opportunities in the arithmetic core by removing the need for expensive multiplications, as well as reducing I/O bandwidth and storage. In this paper, we present an accelerator optimized for binary-weight CNNs that achieves 1.5 TOp/s at 1.2 V on a core area of only 1.33 million gate equivalent (MGE) or 1.9 mm2and with a power dissipation of 895 μW in UMC 65-nm technology at 0.6 V. Our accelerator significantly outperforms the state-of-the-art in terms of energy and area efficiency achieving 61.2 TOp/s/W@0.6 V and 1.1 TOp/s/MGE@1.2 V, respectively.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“YodaNN50, a digital DNN inference accelerator for binary weight networks with ultra-low power budget, achieves 434.8 FPS in high throughput mode for a 9-layer CNN (BinaryConnect51) on CIFAR-10.”
From Accurate deep neural network inference using computational phase-change memory (2020) · cites this paper as [50]
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
Human-level control through deep reinforcement learning
Mnih, V., Kavukcuoglu, K., Silver, D., Rusu, A. A., Veness, J., Bellemare, M. G., Graves, A., Riedmiller, M., Fidjeland, A. K., Ostrovski, G., et al. (2015) · Nature · doi:10.1038/nature14236 · Cited by 30,357 · cited 1×
“On the other hand, sourcing out all CNN computation from IoT end-nodes to data servers is extremely challenging and power consuming, due to the large communication bandwidth required to transmit the data streams.”
BackgroundDeepFace: Closing the Gap to Human-Level Performance in Face Verification
Taigman, Y., Yang, M., Ranzato, M., & Wolf, L. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2014.220 · Cited by 6,542
Han, S., Mao, H., & Dally, W. J. (2015) · arXiv · doi:10.48550/arxiv.1510.00149 · Cited by 4,662
XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks
Rastegari, M., Ordonez, V., Redmon, J., & Farhadi, A. (2016) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-46493-0_32 · Cited by 4,408
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1512.03385 · Cited by 4,196
Eyeriss: An Energy-Efficient Reconfigurable Accelerator for Deep Convolutional Neural Networks
Chen, Y.-H., Krishna, T., Emer, J. S., & Sze, V. (2017) · IEEE Journal of Solid-State Circuits · doi:10.1109/jssc.2016.2616357 · Cited by 3,247 · cited 3×
“For small filter sizes, they can be used to calculate several output channels in parallel or they can be switched-off [41].”
BackgroundMethodLearning Hierarchical Features for Scene Labeling
Farabet, C., Couprie, C., Najman, L., & LeCun, Y. (2013) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2012.231 · Cited by 2,620
Courbariaux, M., Hubara, I., Soudry, D., El-Yaniv, R., & Bengio, Y. (2016) · arXiv · doi:10.48550/arxiv.1602.02830 · Cited by 2,106
(2013) · Cited by 2,056
EIE: efficient inference engine on compressed deep neural network
Han, S., Liu, X., Mao, H., Pu, J., Pedram, A., Horowitz, M. A., & Dally, W. J. (2016) · ACM SIGARCH Computer Architecture News · doi:10.1145/3007787.3001163 · Cited by 1,937
ISAAC: a convolutional neural network accelerator with in-situ analog arithmetic in crossbars
Shafiee, A., Nag, A., Muralimanohar, N., Balasubramonian, R., Strachan, J. P., Hu, M., Williams, R. S., & Srikumar, V. (2016) · ACM SIGARCH Computer Architecture News · doi:10.1145/3007787.3001139 · Cited by 1,536
DianNao: a small-footprint high-throughput accelerator for ubiquitous machine-learning
Chen, T., Du, Z., Sun, N., Wang, J., Wu, C., Chen, Y., & Temam, O. (2014) · Proceedings of the 19th international conference on Architectural support for programming languages and operating systems · doi:10.1145/2541940.2541967 · Cited by 1,297
EIE: Efficient Inference Engine on Compressed Deep Neural Network
Han, S., Liu, X., Mao, H., Pu, J., Pedram, A., Horowitz, M. A., & Dally, W. J. (2016) · 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA) · doi:10.1109/isca.2016.30 · Cited by 1,077
ShiDianNao: shifting vision processing closer to the sensor
Du, Z., Fasthuber, R., Chen, T., Ienne, P., Li, L., Luo, T., Feng, X., Chen, Y., & Temam, O. (2015) · Proceedings of the 42nd Annual International Symposium on Computer Architecture · doi:10.1145/2749469.2750389 · Cited by 933
14.5 Eyeriss: An energy-efficient reconfigurable accelerator for deep convolutional neural networks
Chen, Y.-H., Krishna, T., Emer, J., & Sze, V. (2016) · 2016 IEEE International Solid-State Circuits Conference (ISSCC) · doi:10.1109/isscc.2016.7418007 · Cited by 713
Fixed Point Quantization of Deep Convolutional Networks
Lin, D. D., Talathi, S. S., & Annapureddy, V. S. (2015) · arXiv · doi:10.48550/arxiv.1511.06393 · Cited by 684 · cited 1×
“It was shown that state-of-the-art results can be achieved quantizing the weights and activations of each layer separately [33], while lowering precision down to 2-bit ( − 1, 0, + 1) and increasing the network size [20].”
BackgroundDecomposing a scene into geometric and semantically consistent regions
Gould, S., Fulton, R., & Koller, D. (2009) · 2009 IEEE 12th International Conference on Computer Vision · doi:10.1109/iccv.2009.5459211 · Cited by 672
(2013) · Cited by 651
Cnvlutin: ineffectual-neuron-free deep neural network computing
Albericio, J., Judd, P., Hetherington, T., Aamodt, T., Jerger, N. E., & Moshovos, A. (2016) · ACM SIGARCH Computer Architecture News · doi:10.1145/3007787.3001138 · Cited by 614
Minerva: enabling low-power, highly-accurate deep neural network accelerators
Reagen, B., Whatmough, P., Adolf, R., Rama, S., Lee, H., Lee, S. K., Hernández-Lobato, J. M., Wei, G.-Y., & Brooks, D. (2016) · ACM SIGARCH Computer Architecture News · doi:10.1145/3007787.3001165 · Cited by 446 · cited 1×
“In this work we focus on implementing a CNN inference accelerator for neural networks supporting per-channel scaling and biasing, and implementing binary weights and fixedpoint activation.”
MethodAccelerating Deep Convolutional Neural Networks Using Specialized Hardware
Ovtcharov, K., Ruwase, O., Kim, J.-Y., Fowers, J., Strauß, K., & Chung, E. S. (2015) · Cited by 412
NeuFlow: A runtime reconfigurable dataflow processor for vision
Farabet, C., Martini, B., Corda, B., Akselrod, P., Culurciello, E., & LeCun, Y. (2011) · CVPR 2011 WORKSHOPS · doi:10.1109/cvprw.2011.5981829 · Cited by 369
Graham, B. (2014) · arXiv · doi:10.48550/arxiv.1412.6071 · Cited by 337 · cited 1×
“CNNs are reaching record-breaking accuracy in image recognition on small data sets like MNIST, SVHN, and CIFAR-10 with accuracy rates of 99.79%, 98.31%, and 96.53% [29]–[31].”
BackgroundInfluentialDeep Image: Scaling up Image Recognition
Wu, R., Yan, S., Shan, Y., Dang, Q., & Sun, G. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1501.02876 · Cited by 313
A 240 G-ops/s Mobile Coprocessor for Deep Neural Networks
Gokhale, V., Jin, J., Dundar, A., Martini, B., & Culurciello, E. (2014) · 2014 IEEE Conference on Computer Vision and Pattern Recognition Workshops · doi:10.1109/cvprw.2014.106 · Cited by 300
Hardware-oriented Approximation of Convolutional Neural Networks
Gysel, P., Motamedi, M., & Ghiasi, S. (2016) · arXiv · doi:10.48550/arxiv.1604.03168 · Cited by 287 · cited 2×
“For LeNet and Cifar-10 the additional error introduced by this quantization is less than 0.3% and 2%, respectively, even when the word width has been constrained to 4-bit [21].”
BackgroundMethodInfluentialXNOR-Net: ImageNet Classification Using Binary Convolutional Neural NetworksPreprint
Rastegari, M., Ordóñez, V., Redmon, J., et al. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1603.05279 · Cited by 279
RedEye: analog ConvNet image sensor architecture for continuous mobile vision
LiKamWa, R., Hou, Y., Gao, J., Polansky, M., & Zhong, L. (2016) · ACM SIGARCH Computer Architecture News · doi:10.1145/3007787.3001164 · Cited by 189
Origami: A 803-GOp/s/W Convolutional Network Accelerator
Cavigelli, L., & Benini, L. (2017) · IEEE Transactions on Circuits and Systems for Video Technology · doi:10.1109/tcsvt.2016.2592330 · Cited by 169
Self-citationConvolution engine: balancing efficiency & flexibility in specialized computing
Qadeer, W., Hameed, R., Shacham, O., Venkatesan, P., Kozyrakis, C., & Horowitz, M. A. (2013) · Proceedings of the 40th Annual International Symposium on Computer Architecture · doi:10.1145/2485922.2485925 · Cited by 168
Origami: A Convolutional Network Accelerator
Cavigelli, L., Gschwend, D., Mayer, C., Willi, S., Muheim, B., & Benini, L. (2015) · Proceedings of the 25th edition on Great Lakes Symposium on VLSI · doi:10.1145/2742060.2743766 · Cited by 156
Self-citationSim, J., Park, J.-S., Kim, M., Bae, D., Choi, Y., & Kim, L.-S. (2016) · 2016 IEEE International Solid-State Circuits Conference (ISSCC) · doi:10.1109/isscc.2016.7418008 · Cited by 140
Resiliency of Deep Neural Networks under Quantization
Sung, W., Shin, S., & Hwang, K. (2015) · arXiv · doi:10.48550/arxiv.1511.06488 · Cited by 132
Park, S., Bong, K., Shin, D., Lee, J., Choi, S., & Yoo, H.-J. (2015) · 2015 IEEE International Solid-State Circuits Conference - (ISSCC) Digest of Technical Papers · doi:10.1109/isscc.2015.7062935 · Cited by 108
Generalizing Pooling Functions in Convolutional Neural Networks: Mixed, Gated, and Tree
Lee, C., Gallagher, P. W., & Zhuowen Tu, A. (2015) · arXiv (Cornell University) · doi:10.48550/arxiv.1509.08985 · Cited by 107
Accelerating real-time embedded scene labeling with convolutional networks
Cavigelli, L., Magno, M., & Benini, L. (2015) · Proceedings of the 52nd Annual Design Automation Conference · doi:10.1145/2744769.2744788 · Cited by 106
Self-citationEnergy-efficient ConvNets through approximate computing
Moons, B., De Brabandere, B., Van Gool, L., & Verhelst, M. (2016) · 2016 IEEE Winter Conference on Applications of Computer Vision (WACV) · doi:10.1109/wacv.2016.7477614 · Cited by 98
NeuFlow: Dataflow vision processing system-on-a-chip
Pham, P.-H., Jelaca, D., Farabet, C., Martini, B., LeCun, Y., & Culurciello, E. (2012) · 2012 IEEE 55th International Midwest Symposium on Circuits and Systems (MWSCAS) · doi:10.1109/mwscas.2012.6292202 · Cited by 86
Dialog-based Language Learning
Weston, J. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1604.06045 · Cited by 76
Deep neural networks are robust to weight binarization and other non-linear distortions
Merolla, P. A., Appuswamy, R., Arthur, J. V., Esser, S. K., & Modha, D. S. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1606.01981 · Cited by 74
A ultra-low-energy convolution engine for fast brain-inspired vision in multicore clusters
Conti, F., & Benini, L. (2015) · Design, Automation, and Test in Europe · doi:10.5555/2755753.2755910 · Cited by 64
Self-citationTeman, A., Rossi, D., Meinerzhagen, P., et al. (2016) · ACM Transactions on Design Automation of Electronic Systems · doi:10.1145/2890498 · Cited by 58
Self-citationA Ultra-Low-Energy Convolution Engine for Fast Brain-Inspired Vision in Multicore Clusters
Conti, F., & Benini, L. (2015) · Design, Automation & Test in Europe Conference & Exhibition (DATE), 2015 · doi:10.7873/date.2015.0404 · Cited by 52
Self-citationPark, S., Choi, S., Lee, J., Kim, M., Park, J.-Y., & Yoo, H. (2016) · Digest of technical papers/Digest of technical papers - IEEE International Solid-State Circuits Conference · doi:10.1109/isscc.2016.7418003 · Cited by 29
Binarized Neural Networks on the ImageNet Classification TaskPreprint
Wu, X., Wu, Y., & Zhao, Y. H. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1604.03058 · Cited by 19
A heterogeneous multi-core system-on-chip for energy efficient brain inspired vision
Pullini, A., Conti, F., Rossi, D., et al. (2016) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas.2016.7539213 · Cited by 10
Self-citationMachine outsmarts man in battle of the decade
von Zastrow, M. E. (2016) · The New Scientist · doi:10.1016/s0262-4079(16)30458-4 · Cited by 5
Yoo, H. J., Park, S., Choi, S., et al. (2016) · Cited by 4
(2014) · doi:10.4230/oasics.slate.2021.16 · Cited by 1
Weston, J., Chopra, S., & Bordes, A. (2014) · arXiv · doi:10.48550/arxiv.1410.3916
Research works printed but not linked: 6
- 2014, INS-03510-C1 Datasheet Datasheet of Myriad 2 Vision Processor
- jouppi, 2016, Google Supercharges Machine Learning Tasks with TPU Custom Chip
- chintala, 2016, convnet-benchmarks
- krizhevsky, 2012, Imagenet classification with deep convolutional neural networks, Proc Adv Neural Inf Process Syst, 1106
- lucas, 2016, Yoda
- courbariaux, 2015, BinaryConnect: Training deep neural networks with binary weights during propagations, Proc Adv Neural Inf Process Syst, 3105
Cited by
Model Compression and Hardware Acceleration for Neural Networks: A Comprehensive Survey
Deng, B. L., Li, G., Han, S., Shi, L., & Xie, Y. (2020) · Proceedings of the IEEE · doi:10.1109/jproc.2020.2976475 · Cited by 996
“If only the weights are binarized/ternarized, the costly MAC operations can be reduced to simpler accumulations [274].”
BackgroundESPNetv2: A Light-Weight, Power Efficient, and General Purpose Convolutional Neural Network
Mehta, S., Rastegari, M., Shapiro, L., & Hajishirzi, H. (2019) · 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2019.00941 · Cited by 551
“Low-bit representation: Another approach to improve inference of a pre-trained network is low-bit representation of network weights using quantization [1, 9, 20, 39, 47, 57, 64].”
Accurate deep neural network inference using computational phase-change memory
Joshi, V., Le Gallo, M., Haefeli, S., Boybat, I., Nandakumar, S. R., Piveteau, C., Dazzi, M., Rajendran, B., Sebastian, A., & Eleftheriou, E. (2020) · Nature Communications · doi:10.1038/s41467-020-16108-9 · Cited by 490
PUMA: A Programmable Ultra-efficient Memristor-based Accelerator for Machine Learning Inference
Ankit, A., Hajj, I. E., Chalamalasetti, S. R., Ndu, G., Foltin, M., Williams, R. S., Faraboschi, P., Hwu, W.-M. W., Strachan, J. P., Roy, K., et al. (2019) · Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems · doi:10.1145/3297858.3304049 · Cited by 424
Accurate deep neural network inference using computational phase-change memory
Joshi, V., Gallo, M. L., Haefeli, S., Boybat, I., Nandakumar, S. R., Piveteau, C., Dazzi, M., Rajendran, B., Sebastian, A., & Eleftheriou, E. S. (2020) · RePEc: Research Papers in Economics · doi:10.3929/ethz-b-000416797 · Cited by 379
Metadata
- Authors
- Andri, R., Cavigelli, L., Rossi, D., & Benini, L.
- Affiliations
- ETH Zurich (CH)
- University of Bologna (IT)
- Venue
- IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
- Published
- 2018-01
- Volume
- 37
- Issue
- 1
- Pages
- 48-60
- ISSN
- 0278-0070 (ISSN-L)
- 1937-4151
- OpenAlex
- W2588598812
- Cited by
- 247
- References
- 51
- Citations by year
- 2026: 4 · 2025: 14 · 2024: 11 · 2023: 20 · 2022: 29 · 2021: 24 · 2020: 45 · 2019: 48 · 2018: 38
- FWCI
- 16.02
- Type
- journal article
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Archivio istituzionale della ricerca (Alma Mater Studiorum Università di Bologna) (submittedVersion, other-oa)
- Repository for Publications and Research Data (ETH Zurich) (submittedVersion, other-oa)
- arXiv (Cornell University)
- ETH Zürich Research Collection
- Funders
- Swiss National Science Foundation (162524)
- (MicroLearn: Micropower Deep Learning)
- Armasuisse Science and Technology
- ERC MultiTherman Project (ERC-AdG-291125)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Low-power high-performance VLSI design
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advanced Memory and Neural Computing
- Keywords
- CNN accelerator, ultra-low power, energy efficiency, area efficiency, 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.