Abstract
Machine learning applications for smartphones, such as camera applications, and automatic speech recognition, require a real-time response of deep learning processing. Although one inference processing of deep learning requires more than a billion operations, the development of hardware and deep learning has allowed on-device real-time deep learning processing in smartphones. The performance of smartphone neural processing unit (NPU) has been doubling every year since NPU was integrated into smartphone application processors in 2017. As of 2021, dozens of on-device deep learning applications have been employed in a smartphone. This chapter aims to provide an overview of the history of progress in achieving on-device inference in the smartphone industry. This chapter describes the development and overview of system-on-chip technology and single instruction multiple data architecture such as central processing unit, graphics processing unit, digital signal processor, and NPU, to achieve on-device artificial intelligence used in smartphones. We will also briefly introduce the architectures and features of NPUs widely used in the mobile phone industry. In addition, it introduces strategies such as quantization, pruning, and compression that can increase computational efficiency.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE. 86(11), 2278–2324 (1998)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. Adv. Neural Inf. Proces. Syst. 25, 1097–1105 (2012)
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014)
Galaxy Unpacked January 2021: Official Replay | Samsung, YouTube. https://youtube.be/TD_BZN0bn_U. Accessed 5 Jan 2022
Exynos 990 Mobile Processor: Official Introduction, YouTube. https://www.youtube.com/watch?v=13RgDxD83vl. Accessed 5 Jan 2022
Samsung Exynos 2100 official introduction webpage, Samsung. https://www.samsung.com/semiconductor/minisite/exynos/products/mobileprocessor/exynos-2100/. Accessed 5 Jan 2022
Qualcomm’s Snapdragon 820 processor product brief document, Qualcomm. https://www.qualcomm.com/media/documents/files/snapdragon-820-processor-product-brief.pdf. Accessed 5 Jan 2022
Huawei Kirin 970 AI processor presentation 10/16/17, YouTube. https://www.youtube.com/watch?v=5v2D_QddnFc. Accessed 5 Jan 2022
Qualcomm’s introduction material of Snapdragon 865 5G, Qualcomm. https://www.qualcomm.com/media/documents/files/2019-snapdragon-865-5g-ai-deep-dive-ziad-asghar-jeff-gehlhaar.pdf. Accessed 5 Jan 2022
Qualcomm Redefines Premium with the Flagship Snapdragon 888 5G Mobile Platform, Qualcomm. https://www.qualcomm.com/news/releases/2020/12/02/qualcomm-redefines-premium-flagship-snapdragon-888-5g-mobile-platform. Accessed 5 Jan 2022
Antony, A., Sarika, S.: A review on IoT operating systems. Int. J. Comput. Appl. 975, 8887
Jang, M., Kim, K., Kim, K.: The performance analysis of ARM NEON technology for mobile platforms. In: Proceedings of the 2011 ACM Symposium on Research in Applied Computation, pp. 104–106 (2011)
The A14 Bionic | Apple Special Event 2020, YouTube. https://www.youtube.com/watch?v=5toYNtqbiwg
Han, S., Pool, J., Tran, J., Dally, W.J.: Learning both weights and connections for efficient neural networks. arXiv preprint arXiv:1506.02626 (2015)
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 (2016)
Horowitz, M.: 1.1 computing’s energy problem (and what we can do about it). In: 2014 IEEE International Solid-State Circuits Conference Digest of Technical Papers (ISSCC), pp. 10–14. IEEE (2014)
Overton, M.L.: Numerical computing with IEEE floating point arithmetic. Society for Industrial and Applied Mathematics (2001)
Gholami, A., Kim, S., Dong, Z., Yao, Z., Mahoney, M.W., Keutzer, K.: A survey of quantization methods for efficient neural network inference. arXiv preprint arXiv:2103.13630 (2021)
Tools in Snapdragon Neural Processing Engine SDK Reference Guide, Qualcomm. https://developer.qualcomm.com/sites/default/files/docs/snpe/tools.html. Accessed 5 Jan 2022
September Event 2018 – Apple, YouTube. https://www.youtube.com/watch?v=wFTmQ27S7OQ. Accessed 5 Jan 2022
MediaTek Dimensity 800 introduction, Mediatek. https://www.mediatek.com/products/smartphones/dimensity-800. Accessed 5 Jan 2022
Huawei Kirin 990 5G launch, YouTube. https://www.youtube.com/watch?v=9zkxuxpObKI. Accessed 5 Jan 2022
Song, J., Cho, Y., Park, J.-S., Jang, J.-W., Lee, S., Song, J.-H., Lee, J.-G., Kang, I.: 7.1 an 11.5 tops/w 1024-mac butterfly structure dual-core sparsity-aware neural processing unit in 8nm flagship mobile soc. In: 2019 IEEE International Solid-State Circuits Conference-(ISSCC), pp. 130–132. IEEE (2019)
Liao, H., Tu, J., Xia, J., Zhou, X.: Davinci: A scalable architecture for neural network computing. In: 2019 IEEE Hot Chips 31 Symposium (HCS), pp. 1–44. IEEE Computer Society (2019)
Lin, C.-H., Cheng, C.-C., Tsai, Y.-M., Hung, S.-J., Kuo, Y.-T., Wang, P.H., Tsung, P.-K., et al.: 7.1 a 3.4-to-13.3 tops/w 3.6 tops dual-core deep-learning accelerator for versatile AI applications in 7nm 5g smartphone SOC. In: IEEE International Solid-State Circuits Conference-(ISSCC) 2020, pp. 134–136. IEEE (2020)
Introducing Neon for Armv8-A, Arm. https://developer.arm.com/architectures/instruction-sets/simd-isas/neon/neon-programmers-guide-for-armv8-a/introducing-neon-for-armv8-a/single-page. Accessed 5 Jan 2022 [27]
Codrescu, L.: Architecture of the Hexagon™ 680 DSP for mobile imaging and computer vision. In: 2015 IEEE Hot Chips 27 Symposium (HCS), pp. 1–26. IEEE (2015)
Arm Mali-T760 GPU, Arm. https://developer.arm.com/ip-products/graphics-and-multimedia/mali-gpus/mali-t760-gpu. Accessed 5 Jan 2022
Howard, A.G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., Adam, H.: Mobilenets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861 (2017) [31]
Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., Chen, L.-C.: Mobilenetv2: Inverted residuals and linear bottlenecks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, pp. 4510–4520 (2018)
Howard, A., Sandler, M., Chu, G., Chen, L.-C., Chen, B., Tan, M., Wang, W., et al.: Searching for mobilenetv3. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, pp. 1314–1324 (2019)
Jolivet, R., Lewis, T.J., Gerstner, W.: Generalized integrate-and-fire models of neuronal activity approximate spike trains of a detailed model to a high degree of accuracy. J. Neurophysiol. 92(2), 959–976 (2004)
Peng, P., Mingyu, Y., Weisheng, X.: Running 8-bit dynamic fixed-point convolutional neural network on low-cost arm platforms. In: 2017 Chinese Automation Congress (CAC), pp. 4564–4568. IEEE (2017)
Courbariaux, M., Bengio, Y., David, J.-P.: Binaryconnect: Training deep neural networks with binary weights during propagations. In: Advances in neural information processing systems, pp. 3123–3131 (2015)
Rastegari, M., Ordonez, V., Redmon, J., Farhadi, A.: Xnor-net: Imagenet classification using binary convolutional neural networks. In: European Conference on Computer Vision, pp. 525–542. Springer, Cham (2016) [36]
Li, F., Zhang, B., Liu, B.: Ternary weight networks. arXiv preprint arXiv:1605.04711 (2016)
Hubara, I., Courbariaux, M., Soudry, D., El-Yaniv, R., Bengio, Y.: Quantized neural networks: Training neural networks with low precision weights and activations. J. Mach. Learn. Res. 18(1), 6869–6898 (2017)
Zhou, S., Wu, Y., Ni, Z., Zhou, X., Wen, H., Zou, Y.: Dorefa-net: Training low bitwidth convolutional neural networks with low bitwidth gradients. arXiv preprint arXiv:1606.06160 (2016) [39]
Zhou, A., Yao, A., Guo, Y., Xu, L., Chen, Y.: Incremental network quantization: Towards lossless CNNs with low-precision weights. arXiv preprint arXiv:1702.03044 (2017)
Szymon Migacz: 8-bit inference with TensorRT. NVIDIA. https://on-demand.gputechconf.com/gtc/2017/presentation/s7310-8-bit-inference-with-tensorrt.pdf
Park, H., Kim, D., Kim, S.: TMA: Tera-MACs/W neural hardware inference accelerator with a multiplier-less massive parallel processor. Int. J. Circuit Theory Appl. 49(5), 1399–1409 (2021)
Model optimization in TensorFlow Lite guide: Google. https://www.tensorflow.org/lite/performance/model_optimization. Accessed 5 Jan 2022.
Quantizing a Model in Snapdragon Neural Processing Engine SDK Reference Guide, Qualcomm. https://developer.qualcomm.com/sites/default/files/docs/snpe/model_conversion.html. Accessed 5 Jan 2022.
He, Y., Dong, X., Kang, G., Yanwei, F., Yan, C., Yang, Y.: Asymptotic soft filter pruning for deep convolutional neural networks. IEEE Trans. Cybernet. 50(8), 3594–3604 (2019)
Wu, B., Xu, C., Dai, X., Wan, A., Zhang, P., Yan, Z., Tomizuka, M., Gonzalez, J., Keutzer, K., Vajda, P.: Visual transformers: Token-based image representation and processing for computer vision. arXiv preprint arXiv:2006.03677 (2020)
Dosovitskiy, A., Beyer, L., Kolesnikov, A., Weissenborn, D., Zhai, X., Unterthiner, T., Dehghani, M., et al.: An image is worth 16x16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929 (2020)
Li, Y., Yuan, G., Wen, Y., Hu, E., Evangelidis, G., Tulyakov, S., Wang, Y., Ren, J.: Efficient former: vision transformers at MobileNet speed. arXiv preprint arXiv:2206.01191 (2022)
Acknowledgments
This work was supported by two Institute of Information & Communications Technology Planning & Evaluation (IITP) grants funded by the Korea government (MSIT) (“No.2020-0-00056, To create AI systems that act appropriately and effectively in novel situations that occur in open worlds”).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Park, H., Kim, S. (2023). Overviewing AI-Dedicated Hardware for On-Device AI in Smartphones. In: Mishra, A., Cha, J., Park, H., Kim, S. (eds) Artificial Intelligence and Hardware Accelerators. Springer, Cham. https://doi.org/10.1007/978-3-031-22170-5_4
Download citation
DOI: https://doi.org/10.1007/978-3-031-22170-5_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-22169-9
Online ISBN: 978-3-031-22170-5
eBook Packages: EngineeringEngineering (R0)