Abstract
As neural networks become more complex, the energy required for doing training and inference has resulted in a noticeable shift towards adopting specialized accelerators to meet strict latency and energy constraints that are prevalent in both edge and cloud deployments. These accelerators achieve high performance through parallelism over hundreds of processing elements, and energy efficiency is achieved by reducing data movement and maximizing resource utilization through data reuse. After providing a brief summary of the problems that neural networks have been solving in the domains of Computer Vision, Natural Language Processing, Recommendation Systems and Graph Processing we will discuss how individual layers from each of these different neural networks can be accelerated in an energy-efficient manner. In particular, we focus on design considerations and trade-offs for mapping CNNs, Transformers, and GNNs on AI accelerators that attempt to maximize compute efficiency and minimize energy consumption by reducing the number of access to memory through efficient data reuse.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Raha, A., Kim, S.K., Mathaikutty, D.A., Venkataramanan, G., Mohapatra, D., Sung, R., Brick, C., Chinya, G.N.: Design considerations for edge neural network accelerators: An industry perspective. In: 34th International Conference on VLSI Design and 20th International Conference on Embedded Systems, pp. 328–333 (2021)
Raha, A., Ghosh, S., Mohapatra, D., Mathaikutty, D.A., Sung, R., Brick, C., Raghunathan, V.: Special session: Approximate TinyML systems: Full system approximations for extreme energy-efficiency in intelligent edge devices. In: IEEE 39th International Conference on Computer Design (ICCD), pp. 13–16 (2021)
Sze, V., Chen, Y.H., Yang, T.-J., Emer, J.S.: Efficient processing of deep neural networks: A tutorial and survey. Proc. IEEE 105, 2295–2329 (2017)
Kwon, H., Chatarasi, P., Sarkar, V., Krishna, T., Pellauer, M., Parashar, A.: Maestro: A data-centric approach to understand reuse, performance, and hardware cost of DNN mappings. IEEE Micro 40, 20–29 (2020)
Norrie, T., Patil, N., Yoon, D.H., Kurian, G., Li, S., Laudon, J., Young, C., Jouppi, N.P., Patterson, D.A.: The design process for Google’s training chips: Tpuv2 and tpuv3. IEEE Micro 41, 56–63 (2021)
Jang, J.-W., Lee, S., Kim, D., Park, H., Ardestani, A.S., Choi, Y., Kim, C., Kim, Y., Yu, H., et al.: Sparsity-aware and re-configurable NPU architecture for Samsung flagship mobile soc. In: ACM/IEEE 48th Annual International Symposium on Computer Architecture (ISCA), pp. 15–28, (2021)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: ImageNet classification with deep convolutional neural networks. Commun. ACM 60, 6, 84–90 (2017). https://doi.org/10.1145/3065386
Zhao, Y., Wang, G., Tang, C., Luo, C., Zeng, W., Zha, Z.-J.: A battle of network structures: An empirical study of CNN, transformer, and MLP (2021). arXiv
Meta AI. The latest in machine learning — papers with code. https://paperswithcode.com/
Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., Bengio, Y.: Generative adversarial nets. In: Advances in Neural Information Processing Systems, vol. 27 (2014)
Vaswani, A., Shazeer, N.M., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, L., Polosukhin, I.: Attention is all you need (2017). arXiv
Khan, S., Naseer, M., Hayat, M., Zamir, S.W., Khan, F.S., Shah, M.: Transformers in vision: A survey. ACM Comput. Surv. 54, 1–41 (2021)
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., Minderer, M., Heigold, G., Gelly, S., Uszkoreit, J., Houlsby, N.: An image is worth 16x16 words: Transformers for image recognition at scale (2021). arXiv
Lin, T., Wang, Y., Liu, X., Qiu, X.: A survey of transformers (2021). arXiv
Tolstikhin, I.O., Houlsby, N., Kolesnikov, A., Beyer, L., Zhai, X., Unterthiner, T., Yung, J., Steiner, A., Keysers, D., Uszkoreit, J., et al.: MLP-mixer: An all-MLP architecture for vision. Adv. Neural Inf. Process. Syst. 34, 24261–24272 (2021)
Ko, H., Lee, S., Park, Y., Choi, A.: A survey of recommendation systems: Recommendation models, techniques, and application fields. Electronics 11(1), 141 (2022)
Wu, S., Sun, F., Zhang, W., Cui, B.: Graph neural networks in recommender systems: A survey (2020). arXiv
Zhang, S., Yao, L., Sun, A., Tay, Y.: Deep learning based recommender system: A survey and new perspectives. ACM Comput. Surv. 52(1), 1–38 (2019)
Dong, G., Tang, M., Wang, Z., Gao, J., Guo, S., Cai, L., Gutierrez, R., Campbell, B., Barnes, L.E., Boukhechba, M.L: Graph neural networks in IoT: A survey. ACM Trans. Sensor Netw. (2022). http://nvdla.org/hw/v1/ias/lut-programming.html
Abadal, S., Jain, A., Guirado, R., López-Alonso, J., Alarcón, E.: Computing graph neural networks: A survey from algorithms to accelerators. ACM Comput. Surv. 54(9), 1–38 (2021)
Zhou, J., Cui, G., Hu, S., Zhang, Z., Yang, C., Liu, Z., Wang, L., Li, C., Sun, M.: Graph neural networks: A review of methods and applications. AI Open 1, 57–81 (2020)
NVDLA Open Source Project - LUT programming
Chen, Y.-H., Krishna, T., Emer, J.S., Sze, V.: Eyeriss: An energy-efficient reconfigurable accelerator for deep convolutional neural networks. IEEE J. Solid-State Circuits 52(1), 127–138 (2017)
Chen, Y.-H., Yang, T.J., Emer, J., Sze, V.: Eyeriss v2: A flexible accelerator for emerging deep neural networks on mobile devices. IEEE J. Emerg. Sel. Topics Circuits Syst. 9(2), 292–308 (2019)
Lin, C.-H., Cheng, C.-C., Tsai, Y.-M., Hung, S.-J., Kuo, Y.-T., Wang, P.H., Tsung, P.-K., Hsu, J.-Y., Lai, W.-C., et al.: 7.1 a 3.4-to-13.3tops/w 3.6tops dual-core deep-learning accelerator for versatile AI applications in 7nm 5g smartphone soc. In: IEEE International Solid-State Circuits Conference-(ISSCC), pp. 134–136 (2020)
Jouppi, N.P., Young, C., Patil, N., Patterson, D., Agrawal, G., Bajwa, R., Bates, S., Bhatia, S., Boden, N., Borchers, A., et al.: In-datacenter performance analysis of a tensor processing unit. SIGARCH Comput. Archit. News 45(2), 1–12 (2017)
Qin, E., Samajdar, A., Kwon, H., Nadella, V., Srinivasan, S.M., Das, D., Kaul, B., Krishna, T.: Sigma: A sparse and irregular GEMM accelerator with flexible interconnects for DNN training. In: IEEE International Symposium on High Performance Computer Architecture (HPCA), pp. 58–70 (2020)
NVIDIA. Nvidia ampere architecture (2022). https://www.nvidia.com/en-us/data-center/ampere-architecture/
Parashar, A., Rhu, M., Mukkara, A., Puglielli, A., Venkatesan, R., Khailany, B., Emer, J., Keckler, S.W., Dally, W.J.: Scnn: An accelerator for compressed-sparse convolutional neural networks. In: Proceedings of the 44th Annual International Symposium on Computer Architecture, pp. 27–40 (2017)
Rhu, M., O’Connor, M., Chatterjee, N., Pool, J., Kwon, Y., Keckler, S.W.: Compressing DMA engine: Leveraging activation sparsity for training deep neural networks. In: IEEE International Symposium on High Performance Computer Architecture (HPCA), pp. 78–91 (2018)
IntelⓇ Movidius™ Myriad™ X Vision Processing Unit (VPU). https://www.intel.com/content/www/us/en/products/details/processors/movidius-vpu/movidius-myriad-x.html
Lee, B., Burgess, N.: Some results on Taylor-series function approximation on FPGA. In: The Thirty-Seventh Asilomar Conference on Signals, Systems Computers, vol. 2, pp. 2198–2202 (2003)
Lin, C.-W., Wang, J.-S.: A digital circuit design of hyperbolic tangent sigmoid function for neural networks. In: 2008 IEEE International Symposium on Circuits and Systems (ISCAS), pp. 856–859 (2008)
Leboeuf, K., Namin, A.H., Muscedere, R., Wu, H., Ahmadi, M.: High speed VLSI implementation of the hyperbolic tangent sigmoid function. In: Third International Conference on Convergence and Hybrid Information Technology, vol. 1, pp. 1070–1073 (2008)
Zamanlooy, B., Mirhassani, M.: Efficient VLSI implementation of neural networks with hyperbolic tangent activation function. IEEE Trans. Very Large Scale Integr. Syst. 22(1), 39–48 (2014)
Ioannou, Y.A., Robertson, D.P., Cipolla, R., Criminisi, A.: Deep roots: Improving CNN efficiency with hierarchical filter groups. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 5977–5986 (2017)
Sun, K., Li, M., Liu, D., Wang, J.: Igcv3: Interleaved low-rank group convolutions for efficient deep neural networks. In: BMVC (2018)
Dumoulin, V., Visin, F.: A guide to convolution arithmetic for deep learning (2016). arXiv
Devlin, J., Chang, M.-W., Lee, K., Toutanova, K.: Bert: Pre-training of deep bidirectional transformers for language understanding. In: NAACL (2019)
Radford, A., Wu, J., Child, R., Luan, D., Amodei, D., Sutskever, I.: Language models are unsupervised multitask learners. OpenAI Blog 1, 9 (2019)
Brown, T.B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., et al.: Language models are few-shot learners (2020). arXiv
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., Liu, P.J.: Exploring the limits of transfer learning with a unified text-to-text transformer (2019). arXiv
Liu, Z., Lin, Y., Cao, Y., Hu, H., Wei, Y., Zhang, Z., Lin, S., Guo, B.: Swin transformer: Hierarchical vision transformer using shifted windows. In: IEEE/CVF International Conference on Computer Vision (ICCV), pp. 9992–10002 (2021)
Wang, H., Zhang, Z., Han, S.: SpAtten: Efficient sparse attention architecture with cascade token and head pruning. In: IEEE International Symposium on High-Performance Computer Architecture (HPCA), pp. 97–110 (2021)
He, K., Zhang, X., Ren, S., Sun, J.: “Deep Residual Learning for Image Recognition,” 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778 (2016). https://doi.org/10.1109/CVPR.2016.90
Hamilton, W.L., Ying, R., Leskovec, J.: Inductive representation learning on large graphs (2017). arXiv
Veličković, P., Cucurull, G., Casanova, A., Romero, A., Liò, P., Bengio, Y.: Graph attention networks (2018). arXiv
Yan, M., Deng, L., Hu, X., Liang, L., Feng, Y., Ye, X., Zhang, Z., Fan, D., Xie, Y.: HyGCN: A GCN accelerator with hybrid architecture (2020). arXiv
Stevens, J.R., Das, D., Avancha, S., Kaul, B., Raghunathan, A.: GNNerator: A hardware/software framework for accelerating graph neural networks (2021). arXiv
Liu, Z., Mao, H., Wu, C.Y., Feichtenhofer, C., Darrell, T., Xie, S.: A convnet for the 2020s (2022). arXiv
Susskind, Z., Arden, B., John, L.K., Stockton, P., John, E.B.: Neuro-symbolic AI: An emerging class of AI workloads and their characterization (2021). arXiv
Wang, X., Han, Y., Leung, V.C., Niyato, D., Yan, X., Chen, X.: Convergence of edge computing and deep learning: A comprehensive survey. IEEE Commun. Surv. Tutor. 22(2), 869–904 (2020)
Raha, A., Raghunathan, V.: qLUT: Input-aware quantized table lookup for energy-efficient approximate accelerators. ACM Trans. Embed. Comput. Syst. 16(5s), 1–23 (2017)
Salvator, D., Wu, H., Kulkarni, M., Emmart, N.: Nvidia technical blog: Int4 precision for AI inference (2019). https://www.nvidia.com/en-us/data-center/ampere-architecture/
Choi, J., Venkataramani, S.: Highly accurate deep learning inference with 2-bit precision (2019). https://www.ibm.com/blogs/research/2019/04/2-bit-precision/
Ghosh, S.K., Raha, A., Raghunathan, V.: Approximate inference systems (axis) end-to-end approximations for energy-efficient inference at the edge. In: Proceedings of the ACM/IEEE International Symposium on Low Power Electronics and Design, pp. 7–12 (2020)
Bavikadi, S., Sutradhar, P.R., Khasawneh, K.N., Ganguly, A., Dinakarrao, S.M.P.: A review of in-memory computing architectures for machine learning applications. In: Proceedings of the Great Lakes Symposium on VLSI, pp. 89–94 (2020)
Yu, S., Jiang, H., Huang, S., Peng, X., Lu, A.: Compute-in-memory chips for deep learning: recent trends and prospects. IEEE Circuits Syst. Mag. 21(3), 31–56 (2021)
Bai, L., Zhao, Y., Huang, X.: A CNN accelerator on FPGA using depthwise separable convolution. IEEE Trans. Circuits Syst. II: Express Briefs 65(10), 1415–1419 (2018)
Lu, S., Wang, M., Liang, S., Lin, J., Wang, Z.: Hardware accelerator for multi-head attention and position-wise feed-forward in the transformer. In: IEEE 33rd International System-on-Chip Conference (SOCC), pp. 84–89. IEEE (2020)
Kiningham, K., Re, C., Levis, P.: Grip: A graph neural network accelerator architecture (2020). arXiv
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Raha, A. et al. (2024). Efficient Hardware Acceleration of Emerging Neural Networks for Embedded Machine Learning: An Industry Perspective. In: Pasricha, S., Shafique, M. (eds) Embedded Machine Learning for Cyber-Physical, IoT, and Edge Computing. Springer, Cham. https://doi.org/10.1007/978-3-031-19568-6_5
Download citation
DOI: https://doi.org/10.1007/978-3-031-19568-6_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-19567-9
Online ISBN: 978-3-031-19568-6
eBook Packages: EngineeringEngineering (R0)