Papers
Topics
Authors
Recent
Assistant
AI Research Assistant
Well-researched responses based on relevant abstracts and paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses.
Gemini 2.5 Flash
Gemini 2.5 Flash 134 tok/s
Gemini 2.5 Pro 41 tok/s Pro
GPT-5 Medium 29 tok/s Pro
GPT-5 High 38 tok/s Pro
GPT-4o 105 tok/s Pro
Kimi K2 180 tok/s Pro
GPT OSS 120B 427 tok/s Pro
Claude Sonnet 4.5 37 tok/s Pro
2000 character limit reached

QuATON: Quantization Aware Training of Optical Neurons (2310.03049v2)

Published 4 Oct 2023 in cs.LG, eess.IV, and physics.optics

Abstract: Optical processors, built with "optical neurons", can efficiently perform high-dimensional linear operations at the speed of light. Thus they are a promising avenue to accelerate large-scale linear computations. With the current advances in micro-fabrication, such optical processors can now be 3D fabricated, but with a limited precision. This limitation translates to quantization of learnable parameters in optical neurons, and should be handled during the design of the optical processor in order to avoid a model mismatch. Specifically, optical neurons should be trained or designed within the physical-constraints at a predefined quantized precision level. To address this critical issues we propose a physics-informed quantization-aware training framework. Our approach accounts for physical constraints during the training process, leading to robust designs. We demonstrate that our approach can design state of the art optical processors using diffractive networks for multiple physics based tasks despite quantized learnable parameters. We thus lay the foundation upon which improved optical processors may be 3D fabricated in the future.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (36)
  1. Lin, X. et al. All-optical machine learning using diffractive deep neural networks. \JournalTitleScience 361, DOI: 10.1126/science.aat8084 (2018).
  2. All-optical phase recovery: diffractive computing for quantitative phase imaging. \JournalTitleAdvanced Optical Materials 10, 2200281 (2022).
  3. Herath, K. et al. Differentiable microscopy designs an all optical quantitative phase microscope. \JournalTitlearXiv preprint arXiv:2203.14944 (2022).
  4. Qian, C. et al. Performing optical logic operations by a diffractive neural network. \JournalTitleLight: Science & Applications 9, 59 (2020).
  5. Spatiotemporal diffractive deep neural networks. \JournalTitleOptics Express 32, 1864–1877 (2024).
  6. Yan, T. et al. Fourier-space diffractive deep neural network. \JournalTitlePhysical review letters 123, 023901 (2019).
  7. Shi, J. et al. Multiple-view d 2 nns array: realizing robust 3d object recognition. \JournalTitleOptics Letters 46, 3388–3391 (2021).
  8. Optical fourier techniques for medical image processing and phase contrast imaging. \JournalTitleOptics communications 281, 1876–1888 (2008).
  9. Panchangam, A. et al. Processing of medical images using real-time optical fourier processing. \JournalTitleMedical Physics 28, 22–27 (2001).
  10. Optical image encryption with multistage and multichannel fractional fourier-domain filtering. \JournalTitleOptics Letters 26, 1242–1244 (2001).
  11. Miscuglio, M. et al. Massively parallel amplitude-only fourier neural network. \JournalTitleOptica 7, 1812–1819 (2020).
  12. Haputhanthri, U. et al. From Hours to Seconds: Towards 100x Faster Quantitative Phase Imaging via Differentiable Microscopy. \JournalTitlearXiv preprint (2022).
  13. Physics-Based Learned Design: Optimized Coded-Illumination for Quantitative Phase Imaging. \JournalTitleIEEE Transactions on Computational Imaging 5, DOI: 10.1109/tci.2019.2905434 (2019).
  14. Deep Optics for Single-Shot High-Dynamic-Range Imaging. In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, DOI: 10.1109/CVPR42600.2020.00145 (2020).
  15. Physics-Aware Differentiable Discrete Codesign for Diffractive Optical Neural Networks. In Proceedings of the 41st IEEE/ACM International Conference on Computer-Aided Design, 1–9, DOI: 10.1145/3508352.3549378 (ACM, New York, NY, USA, 2022).
  16. Q8bert: Quantized 8bit bert. In 2019 Fifth Workshop on Energy Efficient Machine Learning and Cognitive Computing-NeurIPS Edition (EMC2-NIPS), 36–39 (IEEE, 2019).
  17. Quantized convolutional neural networks for mobile devices. In Proceedings of the IEEE conference on computer vision and pattern recognition, 4820–4828 (2016).
  18. Learned step size quantization. \JournalTitlearXiv preprint arXiv:1902.08153 (2019).
  19. Liu, Z. et al. Post-training quantization for vision transformer. \JournalTitleAdvances in Neural Information Processing Systems 34, 28092–28103 (2021).
  20. Up or down? adaptive rounding for post-training quantization. In International Conference on Machine Learning, 7197–7206 (PMLR, 2020).
  21. Nahshan, Y. et al. Loss aware post-training quantization. \JournalTitleMachine Learning 110, 3245–3262 (2021).
  22. Yang, J. et al. Quantization Networks. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 7300–7308, DOI: 10.1109/CVPR.2019.00748 (IEEE, 2019).
  23. Gong, R. et al. Differentiable soft quantization: Bridging full-precision and low-bit neural networks. In Proceedings of the IEEE International Conference on Computer Vision, vol. 2019-October, DOI: 10.1109/ICCV.2019.00495 (2019).
  24. Kirtas, M. et al. Quantization-aware training for low precision photonic neural networks. \JournalTitleNeural Networks 155, 561–573 (2022).
  25. Estimating or propagating gradients through stochastic neurons for conditional computation. \JournalTitlearXiv preprint arXiv:1308.3432 (2013).
  26. Gu, J. et al. Roq: A noise-aware quantization scheme towards robust optical neural networks with low-bit controls. In 2020 Design, Automation & Test in Europe Conference & Exhibition (DATE), 1586–1589 (IEEE, 2020).
  27. Goodman, J. W. Introduction to fourier optics. \JournalTitleIntroduction to Fourier optics, 3rd ed., by JW Goodman. Englewood, CO: Roberts & Co. Publishers, 2005 1 (2005).
  28. Categorical reparameterization with gumbel-softmax. In International Conference on Learning Representations (2017).
  29. Image quality assessment: from error visibility to structural similarity. \JournalTitleIEEE transactions on image processing 13, 600–612 (2004).
  30. Deng, L. The mnist database of handwritten digit images for machine learning research. \JournalTitleIEEE Signal Processing Magazine 29, 141–142 (2012).
  31. Krizhevsky, A. Learning multiple layers of features from tiny images (2009).
  32. Tiny imagenet visual recognition challenge (2015).
  33. The BerHu penalty and the grouped effect. \JournalTitlearXiv preprint (2012).
  34. Ratcliffe, J. A. Some Aspects of Diffraction Theory and their Application to the Ionosphere. \JournalTitleReports on Progress in Physics 19, 306, DOI: 10.1088/0034-4885/19/1/306 (1956).
  35. Paszke, A. et al. Automatic differentiation in pytorch (2017).
  36. Adam: A method for stochastic optimization. \JournalTitlearXiv preprint arXiv:1412.6980 (2014).
Citations (1)
View on Semantic Scholar

Summary

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

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

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now
Lightbulb Streamline Icon: https://streamlinehq.com

Continue Learning

We haven't generated follow-up questions for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Generate Now
List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.

User Circle Single Streamline Icon: https://streamlinehq.com Sign Up