Three-dimensional waveguide interconnects for scalable integration of photonic neural networks

Moughames, Johnny; Porté, Xavier; Thiel, Michael; Ulliac, Gwenn; Larger, Laurent; Jacquot, Maxime; Kadic, Muamer; Brunner, Daniel

doi:10.1364/optica.388205

Cited by 111 publications

(83 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A useful comparison assessment on the classification accuracy of Modified National Institute of Standards and Technology (MNIST) data as well as insightful discussions on the fundamental information capacity of the system are also provided. The introduced platform can be leveraged for practical implementation of hybrid photonic-electronic neural networks [130][131][132].…”

Section: Discussionmentioning

confidence: 99%

Meta-optics for spatial optical analog computing

2020

View full text Add to dashboard Cite

Rapidly growing demands for high-performance computing, powerful data processing, and big data necessitate the advent of novel optical devices to perform demanding computing processes effectively. Due to its unprecedented growth in the past two decades, the field of meta-optics offers a viable solution for spatially, spectrally, and/or even temporally sculpting amplitude, phase, polarization, and/or dispersion of optical wavefronts. In this review, we discuss state-of-the-art developments, as well as emerging trends, in computational metastructures as disruptive platforms for spatial optical analog computation. Two fundamental approaches based on general concepts of spatial Fourier transformation and Green’s function (GF) are discussed in detail. Moreover, numerical investigations and experimental demonstrations of computational optical surfaces and metastructures for solving a diverse set of mathematical problems (e.g., integrodifferentiation and convolution equations) necessary for on-demand information processing (e.g., edge detection) are reviewed. Finally, we explore the current challenges and the potential resolutions in computational meta-optics followed by our perspective on future research directions and possible developments in this promising area.

show abstract

Section: Discussionmentioning

confidence: 99%

Meta-optics for spatial optical analog computing

2020

View full text Add to dashboard Cite

show abstract

“…volume. Integration in 2D results in a quadratic scaling of the circuit's area with a network's size [3], cf. Fig.…”

Section: Canonical 3d Photonic Neural Network Architecturementioning

confidence: 99%

“…2(A). In a 3D implementation weights can be stacked, for example, in planes, and for the simplest organization [3], both, the number of required planes and memory-elements per plane scale linearly with the number of neurons. This mitigates the size-scalability roadblock and 3D routing may well be a fundamental prerequisite for scalable and parallel NN chips.…”

Section: Canonical 3d Photonic Neural Network Architecturementioning

confidence: 99%

“…We do not constrain the nature of photonic neurons or the 3D routing strategy. All-optical as well as electro-optical components acting as neurons are possible, and the 3D photonic interconnect can be realized by refractive index modifications in a 3D medium, multiple stacks of diffractive-optics planes [4] as well as complex 3D circuitry of photonic waveguides [3].…”

Section: Canonical 3d Photonic Neural Network Architecturementioning

confidence: 99%

“…An optical waveguide utilizes the principle of total internal reflection, where a medium with a higher refractive index is surrounded by a medium with a lower refractive index. Recently Moughames et al [3] 3D printed such optical waveguides using a Nanoscribe 3D printer and connections in the form of optical splitters realized the dense connectivity between neurons.…”

Section: Photonics and Artificial Intelligence Focusmentioning

confidence: 99%

See 2 more Smart Citations

Optical neural networks: The 3D connection

Dinç

Brunner

2020

Photoniques

Self Cite

View full text Add to dashboard Cite

We motivate a canonical strategy for integrating photonic neural networks (NN) by leveraging 3D printing. Our belief is that a NN’s parallel and dense connectivity is not scalable without 3D integration. 3D additive fabrication complemented with photonic signal transduction can dramatically augment the current capabilities of 2D CMOS and integrated photonics. Here we review some of our recent advances made towards such an architecture.

show abstract

Optical Waveguides Fabricated via Femtosecond Direct Laser Writing: Processes, Materials, and Devices

Cao

Qiu

et al. 2023

Adv Materials Technologies

View full text Add to dashboard Cite

As the most fundamental unit of photonic integration, optical waveguides offer the possibility of developing efficient and practical photonic chips by facilitating the on‐chip integration of devices with different functions. Femtosecond direct laser writing (FsDLW), as a burgeoning 3D microfabrication technology, can realize the rapid formation of arbitrary optical waveguides without any mask inside versatile materials, such as crystalline dielectric crystals, glasses, polymers, and photoresists. This significant material‐independence allows FsDLW to combine different materials and manufacturing processes to implement a cross‐platform solution. This review first describes the different optical loss types and suitable measurement methods for different applicable scenes, then summarizes the two fabrication mechanisms and points out the general direction for adjusting the waveguide properties by fabrication processes. Finally, the application fields and development prospects of different materials are discussed by overviewing the characteristics of different materials and the material selection preferences of different devices. With a panoramic view of the development, it is concluded with insights and perspectives of the future development of optical waveguides and processing technology via FsDLW.

show abstract

Three-dimensional waveguide interconnects for scalable integration of photonic neural networks

Cited by 111 publications

References 39 publications

Meta-optics for spatial optical analog computing

Meta-optics for spatial optical analog computing

Optical neural networks: The 3D connection

Optical Waveguides Fabricated via Femtosecond Direct Laser Writing: Processes, Materials, and Devices

Contact Info

Product

Resources

About