SUANPAN: Scalable Photonic Linear Vector Machine DOI Creative Commons
Xue Feng,

Ziyue Yang,

Chen Li

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Nov. 25, 2024

Abstract Photonic linear operation is a promising approach to handle the extensive vector multiplications in artificial intelligence (AI) techniques due natural bosonic parallelism and high-speed information transmission of photonics. However, there still no universal scalable photonic computing architecture that can be readily merged with existing electronic digital system. Even though it believed maximizing interaction light beams necessary fully utilize tremendous efforts have been made past decades, achieved dimensionality vector-matrix multiplication very limited difficulty scaling up tightly interconnected or highly coupled optical Here, we propose programmable reconfigurable machine perform only inner product two vectors, formed by series independent basic units, while each unit contains one emitter-detector pair. The elemental values processed vectors are prepared time-space domain encoding. Specifically, encoded output duration continuous light-emitter other as position result obtained sum photocurrents all photodetectors. Since among inside, extreme scalability could simply multiplicating without requiring large-scale analog-to-digital converter digital-to-analog arrays. Our encoding inspired traditional Chinese Suanpan abacus, thus denoted SUANPAN. As proof principle, SUANPAN implemented an 8×8 vertical cavity surface emission laser (VCSEL) array MoTe2 two-dimensional material photodetector array. experimental fidelities for randomly generated products over 98% 1-bit, 2-bit, 4-bit 8-bit quantization 95% 8~80 dimensionalities quantization. Two typical AI tasks Ising non-deterministic polynomial-time (NP)-hard optimization problem neural network visual perception performed demonstrate ability architecture. For problem, 1024-dimensional problems successfully solved, which highest dimensional heuristic algorithm. network, competitive classification accuracy 84~88% MNIST (Modified National Institute Standards Technology) handwritten digit dataset. We believe our proposed capable serving fundamental system potential enhance power future various applications.

Language: Английский

A guidance to intelligent metamaterials and metamaterials intelligence DOI Creative Commons
Chao Qian, Ido Kaminer, Hongsheng Chen

et al.

Nature Communications, Journal Year: 2025, Volume and Issue: 16(1)

Published: Jan. 29, 2025

Language: Английский

Citations

11
Integrated Photonic Neural Networks for Equalizing Optical Communication Signals: A Review DOI Creative Commons
Luís C. B. Silva, Pablo Rafael Neves Marciano, Maria Pontes

et al.

Photonics, Journal Year: 2025, Volume and Issue: 12(1), P. 39 - 39

Published: Jan. 4, 2025

The demand for high-capacity communication systems has grown exponentially in recent decades, constituting a technological field constant change. Data transmission at high rates, reaching tens of Gb/s, and over distances that can reach hundreds kilometers, still faces barriers to improvement, such as distortions the transmitted signals. Such include chromatic dispersion, which causes broadening pulse. Therefore, development solutions adequate recovery signals distorted by complex dynamics channel currently constitutes an open problem since, despite existence well-known efficient equalization techniques, these have limitations terms processing time, hardware complexity, especially energy consumption. In this scenario, paper discusses emergence photonic neural networks promising alternative equalizing optical Thus, review focuses on applications, challenges, opportunities implementing integrated scenario signal equalization. main work carried out, ongoing investigations, possibilities new research directions are also addressed. From review, it be concluded perceptron perform slightly better greater than reservoir computing networks, but with lower data rates. It is important emphasize photonics been growing years, so beyond scope address all existing applications networks.

Language: Английский

Citations

2
Ion-modulation optoelectronic neuromorphic devices: mechanisms, characteristics, and applications DOI

Xiaohan Meng,

Runsheng Gao, Xiaojian Zhu

et al.

Journal of Semiconductors, Journal Year: 2025, Volume and Issue: 46(2), P. 021402 - 021402

Published: Feb. 1, 2025

Abstract The traditional von Neumann architecture faces inherent limitations due to the separation of memory and computation, leading high energy consumption, significant latency, reduced operational efficiency. Neuromorphic computing, inspired by human brain, offers a promising alternative integrating computational functions, enabling parallel, high-speed, energy-efficient information processing. Among various neuromorphic technologies, ion-modulated optoelectronic devices have garnered attention their excellent ionic tunability availability multidimensional control strategies. This review provides comprehensive overview recent progress in ion-modulation devices. It elucidates key mechanisms underlying modulation light fields, including ion migration dynamics capture release charge through ions. Furthermore, synthesis active materials properties these are analyzed detail. also highlights application artificial vision systems, other bionic fields. Finally, existing challenges future directions for development discussed, providing critical insights advancing this field.

Language: Английский

Citations

2
Progress on intelligent metasurfaces for signal relay, transmitter, and processor DOI Creative Commons
Chao Qian, Longwei Tian, Hongsheng Chen

et al.

Light Science & Applications, Journal Year: 2025, Volume and Issue: 14(1)

Published: Feb. 25, 2025

Abstract Pursuing higher data rate with limited spectral resources is a longstanding topic that has triggered the fast growth of modern wireless communication techniques. However, massive deployment active nodes to compensate for propagation loss necessitates high hardware expenditure, energy consumption, and maintenance cost, as well complicated network interference issues. Intelligent metasurfaces, composed number subwavelength passive or meta-atoms, have recently found be new paradigm actively reshape environment in green way, distinct from conventional works passively adapt surrounding. In this review, we offer unified perspective on how intelligent metasurfaces can facilitate three manners: signal relay, transmitter, processor. We start by basic modeling channel evolution passive, metasurfaces. Integrated various deep learning algorithms, cater ever-changing environments without human intervention. Then, overview specific experimental advancements using conclude identifying key issues practical implementations surveying directions, such gain knowledge migration.

Language: Английский

Citations

2
Unraveling Chiral Perovskite Spin‐Light Emitting Diode Performance and Magneto‐Chiroptical Properties Relationship Due to the Synergistic Effect DOI Open Access
Yang Li,

Linze Jiang,

Jun Tang

et al.

Advanced Functional Materials, Journal Year: 2025, Volume and Issue: unknown

Published: March 23, 2025

Abstract Chiral hybrid perovskites (CHPs) are very promising for room temperature spin‐light emitting diodes (spin‐LEDs) because of the chiral‐induced spin orbit coupling (CISOC) and helicity‐dependent carrier transport. The chiral‐achiral synergistic method has been recognized as a critical successful pathway developing high‐performance spin‐LEDs. Nonetheless, it remains an absence any studies to demonstrate elucidate relationship between chiral perovskite spin‐LEDs’ performance magneto‐chiroptical properties. Herein, spin‐LEDs being designed fabricated using method. A combination experimental theoretical study is performed systematically toward understanding spin‐related chiroptical properties, instance, lifetimes, CISOC strengths, magnetic transition dipole moments. To tune organic constituent found be decisive degrees circularly polarized electroluminescence (CP‐EL) properties CHPs. This insightful study, first time, unlocks correlation spin‐dependent due effect.

Language: Английский

Citations

1
On-chip Deep Residual Photonic Neural Networks using Optical-electrical Shortcut Connections DOI
Kai Wang, Zuoyin Tang, Yunlong Li

et al.

Optics Letters, Journal Year: 2025, Volume and Issue: 50(3), P. 860 - 860

Published: Jan. 13, 2025

On-chip photonic neural networks (PNNs) have recently emerged as an attractive hardware accelerator for deep learning applications. However, PNNs with higher inference complexity are harder to train due gradient vanishing and exploding problems. In this work, we propose on-chip residual network architecture (Res-PNN), which enables the training of deeper by using optical-electrical shortcut connections. The connection is designed a power splitter, wavelength demultiplexer, photodetectors directly connect input output across optical weight layers. This alleviates problems providing direct path backpropagation, ensuring stable PNNs. proposed Res-PNN achieves classification accuracies 88.4% on CIFAR-10 dataset 80.3% CIFAR-100 dataset. Compared fully connected PNNs, improves accuracy 3.2% 11.3%

Language: Английский

Citations

0
Interfacing Nanophotonics with Deep Neural Networks: AI for Photonic Design and Photonic Implementation of AI DOI Open Access

T. Park,

Sujoy Mondal, Wenshan Cai

et al.

Laser & Photonics Review, Journal Year: 2025, Volume and Issue: unknown

Published: Jan. 29, 2025

Abstract Recent remarkable progress in artificial intelligence (AI) has garnered tremendous attention from researchers, industry leaders, and the general public, who are increasingly aware of AI's growing impact on everyday life. The advancements AI deep learning have also significantly influenced field nanophotonics. On one hand, facilitates data‐driven strategies for optimizing solving forward inverse problems nanophotonic devices. other photonic devices offer promising optical platforms implementing neural networks. This review explores both design implementation AI. Various models their roles introduced, analyzing strengths challenges these methodologies perspective computational cost. Additionally, potential hardware accelerators networks is discussed by presenting a variety capable performing linear nonlinear operations, essential building blocks It believed that bidirectional interactions between nanophotonics will drive coevolution two research fields.

Language: Английский

Citations

0
Advancements in ultrafast photonics: confluence of nonlinear optics and intelligent strategies DOI Creative Commons
Qing Wu,

Liuxing Peng,

Zhifeng Huang

et al.

Light Science & Applications, Journal Year: 2025, Volume and Issue: 14(1)

Published: Feb. 25, 2025

Abstract Automatic mode-locking techniques, the integration of intelligent technologies with nonlinear optics offers promise on-demand control, potentially overcoming inherent limitations traditional ultrafast pulse generation that have predominantly suffered from instability and suboptimality open-loop manual tuning. The advancements in algorithm-driven automatic techniques primarily are explored this review, which also revisits fundamental principles optical absorption, examines evolution categorization conventional techniques. convergence interactions has intricately expanded scope photonics, unveiling considerable potential for innovation catalyzing new waves research breakthroughs photonics characters.

Language: Английский

Citations

0
Computed Tomography Using Meta-Optics DOI
Maksym Zhelyeznyakov, Johannes E. Fröch, Shane Colburn

et al.

ACS Photonics, Journal Year: 2025, Volume and Issue: unknown

Published: March 1, 2025

Language: Английский

Citations

0
Incoherent Optical Neural Networks for Passive and Delay-Free Inference in Natural Light DOI Creative Commons
Rui Chen, Yuan Ma, Zhong Lin Wang

et al.

Photonics, Journal Year: 2025, Volume and Issue: 12(3), P. 278 - 278

Published: March 18, 2025

Optical neural networks are hardware implemented based on physical optics, and they have demonstrated advantages of high speed, low energy consumption, resistance to electromagnetic interference in the field image processing. However, most previous optical were designed for coherent light inputs, which required introduction an electro-optical conversion module before computing device. This significantly hindered inherent speed efficiency computing. In this paper, we propose a diffraction algorithm incoherent mutual intensity propagation, basis, established model network. is completely passive directly performs inference calculations natural light, with detector outputting results, achieving target classification all-optical environment. The proposed was tested MNIST, Fashion-MNIST, ISDD datasets, accuracies 82.32%, 72.48%, 93.05%, respectively, experimental verification showing accuracy error less than 5%. network can achieve delay-free environment, completing good application prospects remote sensing.

Language: Английский

Citations

0