We present subwavelength imaging of amplitude- and phase-encoded objects based on a solid-immersion diffractive processor designed through deep learning. Subwavelength features from the objects are resolved by the collaboration between a jointly-optimized diffractive encoder and decoder pair. We experimentally demonstrated the subwavelength-imaging performance of solid immersion diffractive processors using terahertz radiation and achieved all-optical reconstruction of subwavelength phase features of objects (with linewidths of ~λ/3.4, where λ is the wavelength) by transforming them into magnified intensity images at the output field-of-view. Solid-immersion diffractive processors would provide cost-effective and compact solutions for applications in bioimaging, sensing, and material inspection, among others.
The ever-increasing demand for training and inferring with larger machine-learning models requires more efficient hardware solutions due to limitations such as power dissipation and scalability. Optics is a promising contender for providing lower power computation, since light propagation through a nonabsorbing medium is a lossless operation. However, to carry out useful and efficient computations with light, generating and controlling nonlinearity optically is a necessity that is still elusive. Multimode fibers (MMFs) have been shown that they can provide nonlinear effects with microwatts of average power while maintaining parallelism and low loss. We propose an optical neural network architecture that performs nonlinear optical computation by controlling the propagation of ultrashort pulses in MMF by wavefront shaping. With a surrogate model, optimal sets of parameters are found to program this optical computer for different tasks with minimal utilization of an electronic computer. We show a remarkable decrease of 97% in the number of model parameters, which leads to an overall 99% digital operation reduction compared to an equivalently performing digital neural network. We further demonstrate that a fully optical implementation can also be performed with competitive accuracies.
We will describe MaxwellNet, a machine learning method for analyzing and designing optical systems using Maxwell's equations as the loss function in the optimisation process. We will describe results of the application of MaxwellNet to nanophotonics, nonlinear optics and optical diffraction tomography
Deep neural network trained on physical losses are emerging as promising surrogates of nonlinear numerical solvers. These tools can predict solutions of Maxwell’s equations and compute gradients of output fields with respect to material properties in millisecond times which makes them very attractive for inverse design or inverse scattering applications. Here we demonstrate a neural network able to compute light scattering from inhomogeneous media in the presence of the optical Kerr effect from glass diffusers with a size comparable with the incident wavelength. The weights of the network are dynamically adjusted to take into account the intensity dependent refractive index of the material.
We propose a physics-informed neural network (PINN) as the forward model for tomographic reconstructions of biological samples. We demonstrate that by training this network with the Helmholtz equation as a physical loss, we can predict the scattered field accurately. It will be shown that a pretrained network can be fine-tuned for different samples and used for solving the scattering problem much faster than other numerical solutions. We evaluate our methodology with numerical and experimental results. Our PINNs can be generalized for any forward and inverse scattering problem.
Utilizing light propagation and optical nonlinearities is one of the strategies to accelerate computational tasks in tandem with electrical circuits in an energy-efficient manner. Computing with multimode optical fibers has been demonstrated to be energy efficient due to the high light confinement and multidimensionality. However, these optical nonlinearities have not been programmed for a specific computational task and thus the performance is not optimal. In this study, we demonstrate that the nonlinear transformation in the fiber can be programmed to obtain improved performances on several different machine learning tasks by shaping the wavefront of the information encoding beam.
We experimentally demonstrate phase encoding in SHG with transparent all-dielectric metasurfaces. While a similar task was previously achieved with plasmonic metasurfaces for THG beam shaping, here we obtain three-order-of-magnitude higher generation efficiency without thermal dissipation.
Nanoscale optical integration is nowadays a strategic technological challenge and the ability of generating and manipulating nonlinear optical processes in sub-wavelength volumes is pivotal to realize efficient sensing probes and photonic sources for the next-generation communication technologies. Yet, confining nonlinear processes below the diffraction limit remains a challenging task because phase-matching is not a viable approach at the nanoscale. The localized fields associated to the resonant modes of plasmonic and dielectric nanoantennas offer a route to enhance and control nonlinear processes in highly confined volumes. In my talk I will discuss two nonlinear platforms based on plasmonic and dielectric nanostructures. The first relies on a broken symmetry antenna design, which brings about an efficient second harmonic generation (SHG). We recently applied this concept to an extended array of non-centrosymmetric nanoantennas for sensing applications. I will also show the evidence of a cascaded second-order process in Third Harmonic Generation (THG) in these nanoantennas.
Recently, dielectric nanoantennas emerged as an alternative to plasmonic nanostructures for nanophotonics applications, thanks to their sharp magnetic and electric Mie resonances along with the low ohmic losses in the visible/near-infrared region of the spectrum. I will present our most recent studies on the nonlinear properties of AlGaAs dielectric nanopillars. The strong localized modes along with the broken symmetry in the crystal structure of AlGaAs allow obtaining more than two orders of magnitude higher SHG efficiency with respect to plasmonic nanoantennas with similar spatial footprint and using the same pump power. I will also discuss a few key strategies we recently adopted to optically switch the SHG in these antennas even on the ultrafast time scale. Finally, I will show how to effectively engineer the sum frequency generation via the Mie resonances in these nanoantennas. These results draw a viable blueprint towards room-temperature all optical logic operation at the nanoscale.
We demonstrate photon-pair generation via spontaneous parametric down-conversion (SPDC) from two types of metasurfaces composed by AlGaAs nanocylinders: 1) monolithically fabricated on a selectively oxidized layer of AlAs epitaxially grown on a GaAs wafer; 2) fabricated by reporting the AlGaAs nanostructures on a transparent wafer via wafer bonding. In these samples, we observed SPDC both in back- and forward-scattering configurations, under excitation with a CW pump around 775 nm and single-photon detection on the signal and idler channels. The Bragg modulation of Mie-resonances enables paraxial SPDC, which demonstrates the potential of all-dielectric metasurfaces for quantum applications like on-axis quantum imaging.
All-dielectric optical metasurfaces consist of 2D arrangements of nanoresonators and are of great importance for shaping polarization, phase and amplitude of both linear and harmonic fields. Here, we demonstrate the generation of second harmonic (SH) with zero-order diffraction from nonlinear AlGaAs metasurfaces with spatial period comparable with a pump wavelength in the near-IR. Upon normal incidence of the pump, we demonstrate paraxial SH light into the zero order. SH polarization is effectively controlled via either the meta-atom shape or the pump polarization, with potential applications for on-axis imaging and free-space communication systems.
There are two physical effects that are exploited nowadays to implement flat metalenses requiring 2𝜋-phase excursion, either subwavelength guidance implementing varying propagation delays, or resonant confinement combining two resonances. We compare both approaches and identify possible limitations with the second approach.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.