Informatics Educational Institutions & Programs

Wikipedia

An optical neural network (ONN) is a physical implementation of an artificial neural network with optical components.

Schematic of an optical neural network that functions as a logic gate (above) and its implementation in microwave frequencies (below). The intermediate diffractive metasurfaces function as hidden layers.[1]

Electrochemical vs. optical neural networks

Biological neural networks are electrochemical, while optical neural networks use electromagnetic waves. Optical interfaces to biological neural networks can be created with optogenetics. Biological neural networks use multiple mechanisms to change the state of the neurons. These include short-term and long-term synaptic plasticity. Synaptic plasticity is among the electrophysiological phenomena used to control the efficiency of synaptic transmission, long-term for learning and memory, and short-term for transient changes in synaptic transmission efficiency. Implementing this ideally requires advanced photonic materials.

Desirable properties in photonic materials for optical neural networks include the ability to change their efficiency of transmitting light, based on the intensity of incoming light.

Types

Volume hologram

Early ONNs used a photorefractive volume hologram to interconnect arrays of input neurons to arrays of output with synaptic weights in proportion to the multiplexed hologram's strength.[2] Volume holograms were further multiplexed using spectral hole burning to add one dimension of wavelength to space to achieve four dimensional interconnects of two dimensional arrays of neural inputs and outputs.[3] This led to research on alternative methods using the strength of the optical interconnect to implement neuronal communications.[4]

Silicon photonics

Silicon photonics offers superior speed, but lacks the massive parallelism that free-space optics can deliver.

Free-space optics

Free-space optics offers substantial parallelism. One implementation used phase masks for a handwritten digit classifier.[5] Light passing through stacked 3D-printed phase masks can be read by a photodetector array of ten detectors, each representing a digit class (0 through 9). Although this network can achieve terahertz-speeds, it lacks flexibility, as the phase masks are fabricated for a specific task and cannot be trained.

An alternative method employs a 4F convolutional system. This system uses two lenses to execute the convolution Fourier transforms, enabling passive conversion into the Fourier domain without power consumption or latency. The convolution kernels are task-specific fixed phase masks.[6]

Another technique used kernel tiling to access the parallelism of the 4F system while using a digital micromirror device (DMD) instead of a phase mask. Kernels can be entered into the 4F system and conduct inference.[7]

Typical neural networks are not designed for the 4F systems, using a lower resolution and more channels in their feature maps.

Programmable Optical Array/Analogic Computer

The Programmable Optical Array/Analogic Computer (POAC) was implemented in 2000 based on a modified joint Fourier transform correlator (JTC) and bacteriorhodopsin (BR) as a holographic optical memory. The system offered full parallelism, large array size and the speed of light in an optical convolutional neural network. POAC is a general purpose and programmable array computer that has a wide range of applications including: image processing, pattern recognition, target tracking, real-time video processing, document security, and optical switching.

Hybrid

Taichi is a hybrid ONN that combines the power efficiency and parallelism of optical diffraction and the configurability of optical interference. Taichi offers 13.96 million parameters. Taichi avoids the high error rates that afflict deep (multi-layer) networks by combining clusters of fewer-layer diffractive units with arrays of interferometers for reconfigurable computation. Its encoding protocol divides large network models into sub-models that can be distributed across multiple chiplets in parallel.[8]

Taichi achieved 91.89% accuracy in tests with the Omniglot database. It was also used to generate music Bach and generate images the styles of Van Gogh and Munch.[8]

The developers claimed energy efficiency of up to 160 trillion operations second-1 watt-1 and an area efficiency of 880 trillion multiply-accumulate operations mm-2 or 103 more energy efficient than the NVIDIA H100, and 102 times more energy efficient and 10 times more area efficient than previous ONNs.[8]

Other

ONNs include the Hopfield neural network[9] and the Kohonen self-organizing map with liquid crystal spatial light modulators.[10] Neuromorphic engineering can be used to create neuromorphic photonic systems. Typically, these systems encode information in the networks using spikes, mimicking the functionality of spiking neural networks in optical and photonic hardware.

Other photonic devices have demonstrated neuromorphic functionalities, including vertical-cavity surface-emitting lasers,[11][12] integrated photonic modulators,[13] optoelectronic systems based on superconducting Josephson junctions[14] or systems based on resonant tunnelling diodes.[15]

See also

References

  1. ^ Qian, Chao; Lin, Xiao; Lin, Xiaobin; Xu, Jian; Sun, Yang; Li, Erping; Zhang, Baile; Chen, Hongsheng (2020). "Performing optical logic operations by a diffractive neural network". Light: Science & Applications. 9 (59): 59. Bibcode:2020LSA.....9...59Q. doi:10.1038/s41377-020-0303-2. PMC 7154031. PMID 32337023.
  2. ^ Wagner K, Psaltis D (1988). "Adaptive optical networks using photorefractive crystals". Appl. Opt. 27 (9): 1752–1759. Bibcode:1988ApOpt..27.1752P. doi:10.1364/AO.27.001752. PMID 20531647.
  3. ^ Weverka R, Wagner K, Saffman M (1991). "Fully interconnected, two-dimensional neural arrays using wavelength-multiplexed volume holograms". Optics Letters. 16 (11): 826–828. Bibcode:1991OptL...16..826W. doi:10.1364/OL.16.000826. PMID 19776798.
  4. ^ Wagner K, Psaltis D (1993). "Optical neural networks: an introduction by the feature editors". Appl. Opt. 32 (8): 1261–1263. Bibcode:1993ApOpt..32.1261W. doi:10.1364/AO.32.001261. PMID 20820259.
  5. ^ Lin, Xing; Rivenson, Yair; Yardimci, Nezih T.; Veli, Muhammed; Luo, Yi; Jarrahi, Mona; Ozcan, Aydogan (7 September 2018). "All-optical machine learning using diffractive deep neural networks". Science. 361 (6406): 1004–1008. arXiv:1804.08711. Bibcode:2018Sci...361.1004L. doi:10.1126/science.aat8084. PMID 30049787. S2CID 13753997.
  6. ^ Chang, Julie; Sitzmann, Vincent; Dun, Xiong; Heidrich, Wolfgang; Wetzstein, Gordon (17 August 2018). "Hybrid optical-electronic convolutional neural networks with optimized diffractive optics for image classification". Scientific Reports. 8 (1): 12324. Bibcode:2018NatSR...812324C. doi:10.1038/s41598-018-30619-y. PMC 6098044. PMID 30120316. S2CID 256961403.
  7. ^ Li, Shurui; Miscuglio, Mario; Sorger, Volker J.; Gupta, Puneet (2020). "Channel Tiling for Improved Performance and Accuracy of Optical Neural Network Accelerators". arXiv:2011.07391 [cs.ET].
  8. ^ a b c CHOI, CHARLES Q. (12 April 2024). "AI Chip Trims Energy Budget Back by 99+ Percent - IEEE Spectrum". spectrum.ieee.org. Retrieved 2024-04-17.
  9. ^ Ramachandran R, Gunasekaran N (2000). "Optical Implementation of Two Dimensional Bipolar Hopfield Model Neural Network (Scientific Note)" (PDF). Proceedings-National Science Council Republic of China Part a Physical Science and Engineering. 24 (1): 73–8. Archived from the original (PDF) on 12 October 2004.
  10. ^ Duvillier J, Killinger M, Heggarty K, Yao K, de Bougrenet de la Tocnaye JL (January 1994). "All-optical implementation of a self-organizing map: a preliminary approach". Applied Optics. 33 (2): 258–66. Bibcode:1994ApOpt..33..258D. doi:10.1364/AO.33.000258. PMID 20862015.
  11. ^ Hejda M, Robertson J, Bueno J, Alanis J, Hurtado A (2021-06-01). "Neuromorphic encoding of image pixel data into rate-coded optical spike trains with a photonic VCSEL-neuron". APL Photonics. 6 (6): 060802. Bibcode:2021APLP....6f0802H. doi:10.1063/5.0048674. ISSN 2378-0967.
  12. ^ Robertson J, Hejda M, Bueno J, Hurtado A (April 2020). "Ultrafast optical integration and pattern classification for neuromorphic photonics based on spiking VCSEL neurons". Scientific Reports. 10 (1): 6098. Bibcode:2020NatSR..10.6098R. doi:10.1038/s41598-020-62945-5. PMC 7142074. PMID 32269249.
  13. ^ George JK, Mehrabian A, Amin R, Meng J, de Lima TF, Tait AN, et al. (February 2019). "Neuromorphic photonics with electro-absorption modulators". Optics Express. 27 (4): 5181–5191. arXiv:1809.03545. Bibcode:2019OExpr..27.5181G. doi:10.1364/OE.27.005181. PMID 30876120. S2CID 80625696.
  14. ^ Shainline JM (January 2020). "Fluxonic Processing of Photonic Synapse Events". IEEE Journal of Selected Topics in Quantum Electronics. 26 (1): 1–15. arXiv:1904.02807. Bibcode:2020IJSTQ..2627473S. doi:10.1109/JSTQE.2019.2927473. ISSN 1077-260X. S2CID 102352120.
  15. ^ Romeira B, Javaloyes J, Ironside CN, Figueiredo JM, Balle S, Piro O (September 2013). "Excitability and optical pulse generation in semiconductor lasers driven by resonant tunneling diode photo-detectors". Optics Express. 21 (18): 20931–40. Bibcode:2013OExpr..2120931R. doi:10.1364/OE.21.020931. hdl:10400.1/11954. PMID 24103966. S2CID 480070.
source: http://en.m.wikipedia.org/wiki/Optical_neural_network