Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
139 tokens/sec
GPT-4o
7 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

A Novel Computing Paradigm for MobileNetV3 using Memristor (2402.10512v3)

Published 16 Feb 2024 in cs.AR

Abstract: The increasing computational demands of deep learning models pose significant challenges for edge devices. To address this, we propose a memristor-based circuit design for MobileNetV3, specifically for image classification tasks. Our design leverages the low power consumption and high integration density of memristors, making it suitable for edge computing. The architecture includes optimized memristive convolutional modules, batch normalization modules, activation function modules, global average pooling modules, and fully connected modules. Experimental results on the CIFAR-10 dataset show that our memristor-based MobileNetV3 achieves over 90% accuracy while significantly reducing inference time and energy consumption compared to traditional implementations. This work demonstrates the potential of memristor-based designs for efficient deployment of deep learning models in resource-constrained environments.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (43)
  1. Automatic recognition of ocular surface diseases on smartphone images using densely connected convolutional networks. In 2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), pp.  2786–2789. IEEE, 2021.
  2. Chollet, F. Xception: Deep learning with depthwise separable convolutions. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  1251–1258, 2017.
  3. Chua, L. Memristor-the missing circuit element. IEEE Transactions on circuit theory, 18(5):507–519, 1971.
  4. An introduction to high-level synthesis. IEEE Design & Test of Computers, 26(4):8–17, 2009.
  5. Trends in ai inference energy consumption: Beyond the performance-vs-parameter laws of deep learning. Sustainable Computing: Informatics and Systems, 38:100857, 2023.
  6. Spinnaker2: A large-scale neuromorphic system for event-based and asynchronous machine learning. arXiv preprint arXiv:2401.04491, 2024.
  7. Deep residual learning for image recognition. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  770–778, 2016.
  8. A very high slew-rate cmos operational amplifier. 1989.
  9. Searching for mobilenetv3. In Proceedings of the IEEE/CVF international conference on computer vision, pp.  1314–1324, 2019.
  10. Pointwise convolutional neural networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pp.  984–993, 2018.
  11. Squeezenet: Alexnet-level accuracy with 50x fewer parameters and¡ 0.5 mb model size. arXiv preprint arXiv:1602.07360, 2016.
  12. Ielmini, D. Brain-inspired computing with resistive switching memory (rram): Devices, synapses and neural networks. Microelectronic Engineering, 190:44–53, 2018.
  13. The elusive memristor: properties of basic electrical circuits. European Journal of physics, 30(4):661, 2009.
  14. Neuromemristive circuits for edge computing: A review. IEEE transactions on neural networks and learning systems, 31(1):4–23, 2019.
  15. Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25, 2012.
  16. A native spice implementation of memristor models for simulation of neuromorphic analog signal processing circuits. ACM Transactions on Design Automation of Electronic Systems (TODAES), 27(1):1–24, 2021.
  17. A cmos rectified linear unit operating in weak inversion for memristive neuromorphic circuits. Integration, 87:24–28, 2022.
  18. Analogue signal and image processing with large memristor crossbars. Nature electronics, 1(1):52–59, 2018.
  19. Edge ai: On-demand accelerating deep neural network inference via edge computing. IEEE Transactions on Wireless Communications, 19(1):447–457, 2019.
  20. An adc-less rram-based computing-in-memory macro with binary cnn for efficient edge ai. IEEE Transactions on Circuits and Systems II: Express Briefs, 2023.
  21. Shufflenet v2: Practical guidelines for efficient cnn architecture design. In Proceedings of the European conference on computer vision (ECCV), pp.  116–131, 2018.
  22. A modern primer on processing in memory. In Emerging Computing: From Devices to Systems: Looking Beyond Moore and Von Neumann, pp.  171–243. Springer, 2022.
  23. Automated generation of spice models of memristor-based neural networks from python models. In 2023 International Conference on Industrial Engineering, Applications and Manufacturing (ICIEAM), pp.  895–899. IEEE, 2023.
  24. Cmos implementations of rectified linear activation function. In VLSI Design and Test: 22nd International Symposium, VDAT 2018, Madurai, India, June 28-30, 2018, Revised Selected Papers 22, pp.  121–129. Springer, 2019.
  25. Memristor-based edge computing of shufflenetv2 for image classification. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 40(8):1701–1710, 2020.
  26. The missing memristor found. nature, 453(7191):80–83, 2008.
  27. Xnor-rram: A scalable and parallel resistive synaptic architecture for binary neural networks. In 2018 Design, Automation & Test in Europe Conference & Exhibition (DATE), pp.  1423–1428. IEEE, 2018.
  28. Tuinenga, P. W. SPICE a guide to circuit simulation and analysis using PSpice. Prentice-Hall, Inc., 1995.
  29. Parallel in-memory wireless computing. Nature Electronics, pp.  1–9, 2023.
  30. Memristor-based design of sparse compact convolutional neural network. IEEE Transactions on Network Science and Engineering, 7(3):1431–1440, 2019.
  31. Ckfo: Convolution kernel first operated algorithm with applications in memristor-based convolutional neural network. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 40(8):1640–1647, 2020.
  32. Deep learning incorporating biologically inspired neural dynamics and in-memory computing. Nature Machine Intelligence, 2(6):325–336, 2020.
  33. Mnsim: Simulation platform for memristor-based neuromorphic computing system. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 37(5):1009–1022, 2017.
  34. Memristive crossbar arrays for brain-inspired computing. Nature materials, 18(4):309–323, 2019.
  35. Efficient low-bit neural network with memristor-based reconfigurable circuits. IEEE Transactions on Circuits and Systems II: Express Briefs, 2023.
  36. 16.2 edram-cim: Compute-in-memory design with reconfigurable embedded-dynamic-memory array realizing adaptive data converters and charge-domain computing. In 2021 IEEE International Solid-State Circuits Conference (ISSCC), volume 64, pp.  248–250. IEEE, 2021.
  37. Memristor crossbar deep network implementation based on a convolutional neural network. In 2016 International joint conference on neural networks (IJCNN), pp.  963–970. IEEE, 2016.
  38. Extremely parallel memristor crossbar architecture for convolutional neural network implementation. In 2017 International Joint Conference on Neural Networks (IJCNN), pp.  1696–1703. IEEE, 2017.
  39. Full-circuit implementation of transformer network based on memristor. IEEE Transactions on Circuits and Systems I: Regular Papers, 69(4):1395–1407, 2022.
  40. Fully hardware-implemented memristor convolutional neural network. Nature, 577(7792):641–646, 2020.
  41. Edge learning using a fully integrated neuro-inspired memristor chip. Science, 381(6663):1205–1211, 2023.
  42. Memristive quantized neural networks: A novel approach to accelerate deep learning on-chip. IEEE transactions on cybernetics, 51(4):1875–1887, 2019.
  43. Mnsim 2.0: A behavior-level modeling tool for processing-in-memory architectures. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2023.

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now