Packing Sparse Convolutional Neural Networks for Efficient Systolic Array Implementations: Column Combining Under Joint Optimization
The DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index.
Abstract
This paper describes a novel approach of packing sparse convolutional neural networks into a denser format for efficient implementations using systolic arrays. By combining multiple sparse columns of a convolutional filter matrix into a single dense column stored in the systolic array, the utilization efficiency of the systolic array can be substantially increased (e.g., 8x) due to the increased density of nonzero weights in the resulting packed filter matrix. In combining columns, for each row, all filter weights but the one with the largest magnitude are pruned. The remaining weights are retrained to preserve high accuracy. We study the effectiveness of this joint optimization for both high utilization efficiency and classification accuracy with ASIC and FPGA designs based on efficient bit-serial implementations of multiplier-accumulators. We demonstrate that in mitigating data privacy concerns the retraining can be accomplished with only fractions of the original dataset (e.g., 10% for CIFAR-10). We present analysis and empirical evidence on the superior performance of our column combining approach against prior arts under metrics such as energy efficiency (3x) and inference latency (12x).
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.
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
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155
ImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805
SGDR: Stochastic Gradient Descent with Warm Restarts
Loshchilov, I., & Hutter, F. (2016) · arXiv · doi:10.48550/arxiv.1608.03983 · Cited by 5,309
An overview of gradient descent optimization algorithms
Ruder, S. (2016) · arXiv · doi:10.48550/arxiv.1609.04747 · Cited by 4,962
Han, S., Mao, H., & Dally, W. J. (2015) · arXiv · doi:10.48550/arxiv.1510.00149 · Cited by 4,662
In-Datacenter Performance Analysis of a Tensor Processing Unit
Jouppi, N. P., Young, C., Patil, N., Patterson, D., Agrawal, G., Bajwa, R., Bates, S., Bhatia, S., Boden, N., Borchers, A., et al. (2017) · Proceedings of the 44th Annual International Symposium on Computer Architecture · doi:10.1145/3079856.3080246 · Cited by 4,537 · cited 1×
“These systems, such as the Google TPU [27], the ShiDianNao accelerators [17], and numerous other efforts, including [40], [10], [61], [68], [63], have achieved low power consumption and high throughput.”
BackgroundEyeriss: 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 1×
“These systems, such as the Google TPU [27], the ShiDianNao accelerators [17], and numerous other efforts, including [40], [10], [61], [68], [63], have achieved low power consumption and high throughput.”
BackgroundLearning Efficient Convolutional Networks through Network Slimming
Liu, Z., Li, J., Shen, Z., Huang, G., Yan, S., & Zhang, C. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.298 · Cited by 2,761
Channel Pruning for Accelerating Very Deep Neural Networks
He, Y., Zhang, X., & Sun, J. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.155 · Cited by 2,647 · cited 1×
“This has lead to recent work on structured pruning techniques, which add constraints so that the remaining filter matrix after pruning is still dense [16, 21, 23, 37, 39, 53].”
BackgroundCourbariaux, M., Hubara, I., Soudry, D., El-Yaniv, R., & Bengio, Y. (2016) · arXiv · doi:10.48550/arxiv.1602.02830 · Cited by 2,106
CondenseNet: An Efficient DenseNet Using Learned Group Convolutions
Huang, G., Liu, S., Maaten, L. V. D., & Weinberger, K. Q. (2018) · 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2018.00291 · Cited by 904
To prune, or not to prune: exploring the efficacy of pruning for model compression
Zhu, M., & Gupta, S. (2017) · arXiv · doi:10.48550/arxiv.1710.01878 · Cited by 805 · cited 2×
“For β , we follow a pruning rate schedule in a similar manner to that proposed in [59] and prune up to a target number of nonzeros after column combining for each layer.”
MethodFixed 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 2×
“This quantization has been shown to lead to minimal performance degradation even on challenging datasets [35].”
BackgroundESE: Efficient Speech Recognition Engine with Sparse LSTM on FPGA
Han, S., Kang, J., Mao, H., Hu, Y., Li, X., Li, Y., Xie, D., Luo, H., Yao, S., Wang, Y., et al. (2017) · Proceedings of the 2017 ACM/SIGDA International Symposium on Field-Programmable Gate Arrays · doi:10.1145/3020078.3021745 · Cited by 591
Xception: Deep Learning with Depthwise Separable Convolutions
Chollet, F. (2016) · arXiv · doi:10.48550/arxiv.1610.02357 · Cited by 538
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×
“Over the past several years, there has been extensive work on constructing efficient ASIC and FPGA designs for CNNs which generally consider well studied networks such as LeNet5 [34], AlexNet [30], and VGG-16 [55] including [52, 45, 67, 66, 70, 69, 43, 44, 46, 53, 54, 56, 7, 42].”
BackgroundHardware-oriented Approximation of Convolutional Neural Networks
Gysel, P., Motamedi, M., & Ghiasi, S. (2016) · arXiv · doi:10.48550/arxiv.1604.03168 · Cited by 287 · cited 1×
“We quantize both the inputs and weights to an 8-bit fixed-point representation from the 32-bit float-point representation [36, 20] used during training.”
Method
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
“The weights and psums propagate across Adapted from [14], [257], and [263].”
BackgroundSIGMA: A Sparse and Irregular GEMM Accelerator with Flexible Interconnects for DNN Training
Qin, E., Samajdar, A., Kwon, H., Nadella, V., Srinivasan, S., Das, D., Kaul, B., & Krishna, T. (2020) · 2020 IEEE International Symposium on High Performance Computer Architecture (HPCA) · doi:10.1109/hpca47549.2020.00015 · Cited by 487
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
“The number of address indexes is reduced in the block sparsity, where a single address index is used for a group of weights [238], [239].”
BackgroundAWB-GCN: A Graph Convolutional Network Accelerator with Runtime Workload Rebalancing
Geng, T., Li, A., Shi, R., Wu, C., Wang, T., Li, Y., Haghi, P., Tumeo, A., Che, S., Reinhardt, S., et al. (2020) · 2020 53rd Annual IEEE/ACM International Symposium on Microarchitecture (MICRO) · doi:10.1109/micro50266.2020.00079 · Cited by 301
Ham, T. J., Lee, Y., Seo, S. H., Kim, S., Choi, H., Jung, S. J., & Lee, J. W. (2021) · 2021 ACM/IEEE 48th Annual International Symposium on Computer Architecture (ISCA) · doi:10.1109/isca52012.2021.00060 · Cited by 192
Metadata
- Authors
- Kung, H. T., McDanel, B., & Zhang, S. Q.
- Venue
- Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Systems
- Publisher
- ACM
- Published
- 2019-04-04
- Pages
- 821-834
- OpenAlex
- W2900228909
- Cited by
- 146
- References
- 19
- FWCI
- 8.60
- Type
- conference paper
- Language
- en
- Open access
- closed
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.