-
Convergence and scaling of Boolean-weight optimization for hardware reservoirs
Authors:
Louis Andreoli,
Stéphane Chrétien,
Xavier Porte,
Daniel Brunner
Abstract:
Hardware implementation of neural network are an essential step to implement next generation efficient and powerful artificial intelligence solutions.
Besides the realization of a parallel, efficient and scalable hardware architecture, the optimization of the system's extremely large parameter space with sampling-efficient approaches is essential.
Here, we analytically derive the scaling laws…
▽ More
Hardware implementation of neural network are an essential step to implement next generation efficient and powerful artificial intelligence solutions.
Besides the realization of a parallel, efficient and scalable hardware architecture, the optimization of the system's extremely large parameter space with sampling-efficient approaches is essential.
Here, we analytically derive the scaling laws for highly efficient Coordinate Descent applied to optimizing the readout layer of a random recurrently connection neural network, a reservoir.
We demonstrate that the convergence is exponential and scales linear with the network's number of neurons.
Our results perfectly reproduce the convergence and scaling of a large-scale photonic reservoir implemented in a proof-of-concept experiment.
Our work therefore provides a solid foundation for such optimization in hardware networks, and identifies future directions that are promising for optimizing convergence speed during learning leveraging measures of a neural network's amplitude statistics and the weight update rule.
△ Less
Submitted 13 May, 2023;
originally announced May 2023.
-
Deterministic and stochastic coarsening control in optically-addressed spatial light modulators subject to optical feedback
Authors:
Vladimir V. Semenov,
Xavier Porte,
Laurent Larger,
Daniel Brunner
Abstract:
Phase separation accompanied by further domain growth and coarsening is a phenomenon common to a broad variety of dynamical systems. In this connection, controlling such processes represents a relevant interdisciplinary problem. Using methods of numerical modelling, we demonstrate two approaches for the coarsening control in bistable systems based on the example of a spatially-extended model descr…
▽ More
Phase separation accompanied by further domain growth and coarsening is a phenomenon common to a broad variety of dynamical systems. In this connection, controlling such processes represents a relevant interdisciplinary problem. Using methods of numerical modelling, we demonstrate two approaches for the coarsening control in bistable systems based on the example of a spatially-extended model describing an optically-addressed spatial light modulator with two color illumination subject to optical feedback. The first method implies varying system parameters such that the system evolves as the pitchfork or saddle-node normal forms. The second method leverages noise whose intensity is used as an additional system parameter. Both, deterministic and stochastic schemes allow to control the direction and speed of the fronts separating spatial domains. The considered stochastic control represents a particular case of the noise-sustained front propagation in bistable systems and involves the properties of the optical system under study. In contrast, the proposed deterministic control technique can be applied to bistable systems of different nature.
△ Less
Submitted 27 February, 2023;
originally announced February 2023.
-
Additive 3D photonic integration that is CMOS compatible
Authors:
Adria Grabulosa,
Johnny Moughames,
Xavier Porte,
Muamer Kadic,
Daniel Brunner
Abstract:
Today, continued miniaturization in electronic integrated circuits (ICs) appears to have reached its fundamental limit. At the same time, energy consumption due by communication becomes the dominant limitation in high performance electronic ICs for computing, and modern computing concepts such a neural networks further amplify the challenge. Photonic communication is a promising strategy to addres…
▽ More
Today, continued miniaturization in electronic integrated circuits (ICs) appears to have reached its fundamental limit. At the same time, energy consumption due by communication becomes the dominant limitation in high performance electronic ICs for computing, and modern computing concepts such a neural networks further amplify the challenge. Photonic communication is a promising strategy to address the second, while adding a third dimension to the predominantly two dimensional integrated circuits appears the most promising future strategy for further IC architecture improvement. Crucial for efficient electronic-photonic co-integration is CMOS compatibility. Here, we review our latest results obtained in the FEMTO-ST RENATECH facilities on using additive photo-induced polymerization of a standard photo-resin for truly 3D photonic integration according to these principles. Based on one- and two-photon polymerization and combined with direct-laser writing, we 3D-printed air- and polymer-cladded photonic waveguides. An important application of such circuits are the interconnects of optical neural networks, where 3D integration enables scalability in terms of network size versus its geometric dimensions. In particular via \emph{flash}-TPP, a fabrication process combining blanket one- and high-resolution two-photon polymerization, we demonstrated polymer-cladded step-index waveguides with up to 6~mm length, low insertion ($\sim$0.26~dB) and propagation ($\sim$1.3~dB/mm) losses, realized broadband and low loss ($\sim$0.06~dB splitting losses) adiabatic 1 to M couplers as well as tightly confining air-cladded waveguides for denser integration. By stably printing such integrated photonic circuits on standard semiconductor samples, we show the concept's CMOS compatibility. With this, we lay out a promising, future avenue for scalable integration of hybrid photonic and electronic components.
△ Less
Submitted 3 January, 2023;
originally announced January 2023.
-
(3+1)D-printed adiabatic 1-to-N broadband couplers
Authors:
Adrià Grabulosa,
Xavier Porte,
Erik Jung,
Johnny Moughames,
Muamer Kadic,
Daniel Brunner
Abstract:
We report single-mode 3D optical couplers leveraging adiabatic power transfer towards up to 4 output ports. We use the CMOS compatible additive (3+1)D \emph{flash}-TPP printing for fast and scalable fabrication. Coupling optical losses of such devices are reduced below $\sim$~0.06~dB by tailoring the coupling and waveguides geometry, and we demonstrate almost octave-spanning broadband functionalit…
▽ More
We report single-mode 3D optical couplers leveraging adiabatic power transfer towards up to 4 output ports. We use the CMOS compatible additive (3+1)D \emph{flash}-TPP printing for fast and scalable fabrication. Coupling optical losses of such devices are reduced below $\sim$~0.06~dB by tailoring the coupling and waveguides geometry, and we demonstrate almost octave-spanning broadband functionality from 520~nm to 980~nm.
△ Less
Submitted 20 August, 2022;
originally announced August 2022.
-
Computational metrics and parameters of an injection-locked large area semiconductor laser for neural network computing
Authors:
Anas Skalli,
Xavier Porte,
Nasibeh Haghighi,
Stephan Reitzenstein,
James A. Lott,
D. Brunner
Abstract:
Artificial neural networks have become a staple computing technique in many fields. Yet, they present fundamental differences with classical computing hardware in the way they process information. Photonic implementations of neural network architectures potentially offer fundamental advantages over their electronic counterparts in terms of speed, processing parallelism, scalability and energy effi…
▽ More
Artificial neural networks have become a staple computing technique in many fields. Yet, they present fundamental differences with classical computing hardware in the way they process information. Photonic implementations of neural network architectures potentially offer fundamental advantages over their electronic counterparts in terms of speed, processing parallelism, scalability and energy efficiency. Scalable and high performance photonic neural networks (PNNs) have been demonstrated, yet they remain scarce. In this work, we study the performance of such a scalable, fully parallel and autonomous PNN based on a large area vertical-cavity surface-emitting laser (LA-VCSEL). We show how the performance varies with different physical parameters, namely, injection wavelength, injection power, and bias current. Furthermore, we link these physical parameters to the general computational measures of consistency and dimensionality. We present a general method of gauging dimensionality in high dimensional nonlinear systems subject to noise, which could be applied to many systems in the context of neuromorphic computing. Our work will inform future implementations of spatially multiplexed VCSEL PNNs.
△ Less
Submitted 16 December, 2021;
originally announced December 2021.
-
Photonic neuromorphic computing using vertical cavity semiconductor lasers
Authors:
Anas Skalli,
Joshua Robertson,
Dafydd Owen-Newns,
Matej Hejda,
Xavier Porte,
Stephan Reitzenstein,
Antonio Hurtado,
D. Brunner
Abstract:
Photonic realizations of neural network computing hardware are a promising approach to enable future scalability of neuromorphic computing. In this review we provide an overview on vertical-cavity surface-emitting lasers (VCSELs) and how these high-performance electro-optical components either implement or are combined with additional photonic hardware to demonstrate points (i-iii). In the neurmor…
▽ More
Photonic realizations of neural network computing hardware are a promising approach to enable future scalability of neuromorphic computing. In this review we provide an overview on vertical-cavity surface-emitting lasers (VCSELs) and how these high-performance electro-optical components either implement or are combined with additional photonic hardware to demonstrate points (i-iii). In the neurmorphic photonics' context, VCSELs are of exceptional interest as they are compatible with CMOS fabrication, readily achieve 30\% wall-plug efficiency and >30~GHz modulation bandwidth and hence are highly energy efficient and ultra-fast. Crucially, they react highly nonlinear to optical injection as well as to electrical modulation, making them highly suitable as all-optical as well as electro-optical photonic neurons. Their optical cavities are wavelength-limited, and standard semiconductor growth and lithography enables non-classical cavity configurations and geometries. This enables excitable VCSELs (i.e. spiking VCSELs) to finely control their temporal and spatial coherence, to unlock Terahertz bandwidths through spin-flip effects, and even to leverage cavity quantum electrodynamics to further boost their efficiency. Finally, as VCSEL arrays they are compatible with standard 2D photonic integration, but their emission vertical to the substrate makes them ideally suited for scalable integrated networks leveraging 3D photonic waveguides. Here, we discuss the implementation of spatially as well as temporally multiplexed VCSEL neural networks and reservoirs, computation on the basis of excitable VCSELs as photonic spiking neurons, as well as concepts and advances in the fabrication of VCSELs and microlasers.
△ Less
Submitted 15 December, 2021;
originally announced December 2021.
-
Combining one and two photon polymerization for accelerated high performance (3+1)D photonic integration
Authors:
Adria Grabulosa,
Johnny Moughames,
Xavier Porte,
Daniel Brunner
Abstract:
Dense and efficient circuits with component sizes approaching the physical limit is the hallmark of high performance integration. However, current fabrication technology is mostly constraint to 2D lithography, and thermal energy dissipation induced by switching electronic signal lines presents a fundamental challenge for truly 3D electronic integration. Photonics reduces this problem, and direct l…
▽ More
Dense and efficient circuits with component sizes approaching the physical limit is the hallmark of high performance integration. However, current fabrication technology is mostly constraint to 2D lithography, and thermal energy dissipation induced by switching electronic signal lines presents a fundamental challenge for truly 3D electronic integration. Photonics reduces this problem, and direct laser writing of a photoresin is a promising high-resolution tool for 3D photonic integration. Here, we combine one and two-photon polymerization (TPP) for waveguide integration for the first time, dramatically accelerating the fabrication process and increasing optical confinement. 3D additive printing is based on femtosecond TPP, while blanket irradiation with a UV lamp induces one-photon polymerization (OPP) throughout the entire 3D chip. We locally and dynamically adjust writing conditions to implement (3+1)D \emph{flash}-TPP: waveguide cores are printed with a small distance between neighboring writing voxels to ensure smooth interfaces, mechanical support structures are printed at maximal distance between the voxels to speed up the process. Finally, the entire chip's \emph{passive} volume not part of waveguide cores or mechanical support is polymerized in a single instance by UV blanket irradiation. This decouples fabrication time from the passive volume's size. We succeed in printing vertical single-mode waveguides of 6~mm length that reach up to NA = 0.16. Noteworthy, we achieve exceptionally low -0.26 dB injection and very low propagation losses of -1.36 dB/mm, which is within one order of magnitude of standard integrated silicon photonics. Finally, the optical performance of our waveguides does not deteriorate during 120 days within our measurement uncertainty.
△ Less
Submitted 14 December, 2021;
originally announced December 2021.
-
Two-color optically-addressed spatial light modulator as generic spatio-temporal systems
Authors:
Vladimir Semenov,
Xavier Porte,
Ibrahim Abdulhalim,
Laurent Larger,
Daniel Brunner
Abstract:
Nonlinear spatio-temporal systems are the basis for countless physical phenomena in such diverse fields as ecology, optics, electronics and neuroscience. The canonical approach to unify models originating from different fields is the normal form description, which determines the generic dynamical aspects and different bifurcation scenarios. Realizing different types of dynamical systems via one ex…
▽ More
Nonlinear spatio-temporal systems are the basis for countless physical phenomena in such diverse fields as ecology, optics, electronics and neuroscience. The canonical approach to unify models originating from different fields is the normal form description, which determines the generic dynamical aspects and different bifurcation scenarios. Realizing different types of dynamical systems via one experimental platform that enables continuous transition between normal forms through tuning accessible system parameters is therefore highly relevant. Here, we show that a transmissive, optically-addressed spatial light modulator under coherent optical illumination and optical feedback coupling allows tuning between pitchfork, transcritical and saddle-node bifurcations of steady states. We demonstrate this by analytically deriving the system's normal form equations and confirm these results via extensive numerical simulations. Our model describes a nematic liquid crystal device using nano-dimensional dichalcogenide (a-As$_2$S$_3$) glassy thin-films as photo sensors and alignment layers, and we use device parameters obtained from experimental characterization. Optical coupling, for example using diffraction, holography or integrated unitary maps allow implementing a variety of system topologies of technological relevance for neural networks and potentially XY-Hamiltonian models with ultra low energy consumption.
△ Less
Submitted 5 November, 2021; v1 submitted 2 November, 2021;
originally announced November 2021.
-
Direct (3+1)D laser writing of graded-index optical elements
Authors:
Xavier Porte,
Niyazi Ulas Dinc,
Johnny Moughames,
Giulia Panusa,
Caroline Juliano,
Muamer Kadic,
Christophe Moser,
Daniel Brunner,
Demetri Psaltis
Abstract:
We propose single-step additive fabrication of graded-index optical elements by introducing the light exposure as the additional dimension to three-dimensional (3D) laser writing, hence (3+1)D writing. We use a commercial printer and photoresist to realize the proposed single step fabrication method that can be swiftly adopted for research and engineering. After presenting the characterization of…
▽ More
We propose single-step additive fabrication of graded-index optical elements by introducing the light exposure as the additional dimension to three-dimensional (3D) laser writing, hence (3+1)D writing. We use a commercial printer and photoresist to realize the proposed single step fabrication method that can be swiftly adopted for research and engineering. After presenting the characterization of the graded-index profiles via basic structures, we demonstrate two different optical devices: volume holograms that are superimposed using angular and peristrophic multiplexing, and optical waveguides with well-defined refractive index profiles. In the latter, we precisely control the propagating modes via tuning the (3+1)D-printed waveguide parameters and report step-index and graded-index core-cladding transitions.
△ Less
Submitted 21 September, 2021; v1 submitted 3 June, 2021;
originally announced June 2021.
-
Optical pumping of quantum dot micropillar lasers
Authors:
L. Andreoli,
X. Porte,
T. Heuser,
J. Grosse,
B. Moeglen-Paget,
L. Furfaro,
S. Reitzenstein,
D. Brunner
Abstract:
Arrays of quantum dot micropillar lasers are an attractive technology platform for various applications in the wider field of nanophotonics. Of particular interest is the potential efficiency enhancement as consequence of cavity quantum electrodynamics effects which makes them prime candidates for next generation photonic neurons in neural network hardware. However, in particular for optical pumpi…
▽ More
Arrays of quantum dot micropillar lasers are an attractive technology platform for various applications in the wider field of nanophotonics. Of particular interest is the potential efficiency enhancement as consequence of cavity quantum electrodynamics effects which makes them prime candidates for next generation photonic neurons in neural network hardware. However, in particular for optical pumping their power-conversion efficiency can be very low. Here we perform an in-depth experimental analysis of quantum dot microlasers and investigate their input-output relationship over a wide range of optical pumping conditions. We find that the current energy efficiency limitation is caused by disadvantageous optical pumping concepts and by a low exciton conversion efficiency. Our results indicate that for non-resonant pumping into the GaAs matrix (wetting layer), 3.4% (0.6%) of the optical pump is converted into lasing-relevant excitons, and of those only 2% (0.75%) provide gain to the lasing transition. Based on our findings we propose to improve the pumping efficiency by orders of magnitude by increasing the aluminium content of the AlGaAs/GaAs mirror pairs in the upper Bragg reflector.
△ Less
Submitted 4 March, 2021; v1 submitted 15 February, 2021;
originally announced February 2021.
-
A complete, parallel and autonomous photonic neural network in a semiconductor multimode laser
Authors:
Xavier Porte,
Anas Skalli,
Nasibeh Haghighi,
Stephan Reitzenstein,
James A. Lott,
Daniel Brunner
Abstract:
Neural networks are one of the disruptive computing concepts of our time. However, they fundamentally differ from classical, algorithmic computing in a number of fundamental aspects. These differences result in equally fundamental, severe and relevant challenges for neural network computing using current computing substrates. Neural networks urge for parallelism across the entire processor and for…
▽ More
Neural networks are one of the disruptive computing concepts of our time. However, they fundamentally differ from classical, algorithmic computing in a number of fundamental aspects. These differences result in equally fundamental, severe and relevant challenges for neural network computing using current computing substrates. Neural networks urge for parallelism across the entire processor and for a co-location of memory and arithmetic, i.e. beyond von Neumann architectures. Parallelism in particular made photonics a highly promising platform, yet until now scalable and integratable concepts are scarce. Here, we demonstrate for the first time how a fully parallel and fully implemented photonic neural network can be realized using spatially distributed modes of an efficient and fast semiconductor laser. Importantly, all neural network connections are realized in hardware, and our processor produces results without pre- or post-processing. 130+ nodes are implemented in a large-area vertical cavity surface emitting laser, input and output weights are realized via the complex transmission matrix of a multimode fiber and a digital micro-mirror array, respectively. We train the readout weights to perform 2-bit header recognition, a 2-bit XOR and 2-bit digital analog conversion, and obtain < 0.9 10^-3 and 2.9 10^-2 error rates for digit recognition and XOR, respectively. Finally, the digital analog conversion can be realized with a standard deviation of only 5.4 10^-2. Our system is scalable to much larger sizes and to bandwidths in excess of 20 GHz.
△ Less
Submitted 21 December, 2020;
originally announced December 2020.
-
Boolean learning under noise-perturbations in hardware neural networks
Authors:
Louis Andreoli,
Xavier Porte,
Stéphane Chrétien,
Maxime Jacquot,
Laurent Larger,
Daniel Brunner
Abstract:
A high efficiency hardware integration of neural networks benefits from realizing nonlinearity, network connectivity and learning fully in a physical substrate. Multiple systems have recently implemented some or all of these operations, yet the focus was placed on addressing technological challenges. Fundamental questions regarding learning in hardware neural networks remain largely unexplored. No…
▽ More
A high efficiency hardware integration of neural networks benefits from realizing nonlinearity, network connectivity and learning fully in a physical substrate. Multiple systems have recently implemented some or all of these operations, yet the focus was placed on addressing technological challenges. Fundamental questions regarding learning in hardware neural networks remain largely unexplored. Noise in particular is unavoidable in such architectures, and here we investigate its interaction with a learning algorithm using an opto-electronic recurrent neural network. We find that noise strongly modifies the system's path during convergence, and surprisingly fully decorrelates the final readout weight matrices. This highlights the importance of understanding architecture, noise and learning algorithm as interacting players, and therefore identifies the need for mathematical tools for noisy, analogue system optimization.
△ Less
Submitted 25 June, 2021; v1 submitted 27 March, 2020;
originally announced March 2020.
-
Three dimensional waveguide-interconnects for scalable integration of photonic neural networks
Authors:
Johnny Moughames,
Xavier Porte,
Michael Thiel,
Gwenn Ulliac,
Maxime Jacquot,
Laurent Larger,
Muamer Kadic,
Daniel Brunner
Abstract:
Photonic waveguides are prime candidates for integrated and parallel photonic interconnects. Such interconnects correspond to large-scale vector matrix products, which are at the heart of neural network computation. However, parallel interconnect circuits realized in two dimensions, for example by lithography, are strongly limited in size due to disadvantageous scaling. We use three dimensional (3…
▽ More
Photonic waveguides are prime candidates for integrated and parallel photonic interconnects. Such interconnects correspond to large-scale vector matrix products, which are at the heart of neural network computation. However, parallel interconnect circuits realized in two dimensions, for example by lithography, are strongly limited in size due to disadvantageous scaling. We use three dimensional (3D) printed photonic waveguides to overcome this limitation. 3D optical-couplers with fractal topology efficiently connect large numbers of input and output channels, and we show that the substrate's footprint area scales linearly. Going beyond simple couplers, we introduce functional circuits for discrete spatial filters identical to those used in deep convolutional neural networks.
△ Less
Submitted 9 January, 2020; v1 submitted 17 December, 2019;
originally announced December 2019.
-
Reservoir-size dependent learning in analogue neural networks
Authors:
Xavier Porte,
Louis Andreoli,
Maxime Jacquot,
Laurent Larger,
Daniel Brunner
Abstract:
The implementation of artificial neural networks in hardware substrates is a major interdisciplinary enterprise. Well suited candidates for physical implementations must combine nonlinear neurons with dedicated and efficient hardware solutions for both connectivity and training. Reservoir computing addresses the problems related with the network connectivity and training in an elegant and efficien…
▽ More
The implementation of artificial neural networks in hardware substrates is a major interdisciplinary enterprise. Well suited candidates for physical implementations must combine nonlinear neurons with dedicated and efficient hardware solutions for both connectivity and training. Reservoir computing addresses the problems related with the network connectivity and training in an elegant and efficient way. However, important questions regarding impact of reservoir size and learning routines on the convergence-speed during learning remain unaddressed. Here, we study in detail the learning process of a recently demonstrated photonic neural network based on a reservoir. We use a greedy algorithm to train our neural network for the task of chaotic signals prediction and analyze the learning-error landscape. Our results unveil fundamental properties of the system's optimization hyperspace. Particularly, we determine the convergence speed of learning as a function of reservoir size and find exceptional, close to linear scaling. This linear dependence, together with our parallel diffractive coupling, represent optimal scaling conditions for our photonic neural network scheme.
△ Less
Submitted 23 July, 2019;
originally announced August 2019.
-
Fundamental aspects of noise in analog-hardware neural networks
Authors:
Nadezhda Semenova,
Xavier Porte,
Louis Andreoli,
Maxime Jacquot,
Laurent Larger,
Daniel Brunner
Abstract:
We study and analyze the fundamental aspects of noise propagation in recurrent as well as deep, multi-layer networks. The main focus of our study are neural networks in analogue hardware, yet the methodology provides insight for networks in general. The system under study consists of noisy linear nodes, and we investigate the signal-to-noise ratio at the network's outputs which is the upper limit…
▽ More
We study and analyze the fundamental aspects of noise propagation in recurrent as well as deep, multi-layer networks. The main focus of our study are neural networks in analogue hardware, yet the methodology provides insight for networks in general. The system under study consists of noisy linear nodes, and we investigate the signal-to-noise ratio at the network's outputs which is the upper limit to such a system's computing accuracy. We consider additive and multiplicative noise which can be purely local as well as correlated across populations of neurons. This covers the chief internal-perturbations of hardware networks and noise amplitudes were obtained from a physically implemented recurrent neural network and therefore correspond to a real-world system. Analytic solutions agree exceptionally well with numerical data, enabling clear identification of the most critical components and aspects for noise management. Focusing on linear nodes isolates the impact of network connections and allows us to derive strategies for mitigating noise. Our work is the starting point in addressing this aspect of analogue neural networks, and our results identify notoriously sensitive points while simultaneously highlighting the robustness of such computational systems.
△ Less
Submitted 21 July, 2019;
originally announced July 2019.
-
Diffractive coupling for photonic networks: how big can we go?
Authors:
Sheler Maktoobi,
Luc Froehly,
Louis Andreoli,
Xavier Porte,
Maxime Jacquot,
Laurent Larger,
Daniel Brunner
Abstract:
Photonic networks are considered a promising substrate for high-performance future computing systems. Compared to electronics, photonics has significant advantages for a fully parallel implementation of networks. A promising approach for parallel large-scale photonic networks is realizing the connections using diffraction. Here, we characterize the scalability of such diffractive coupling in great…
▽ More
Photonic networks are considered a promising substrate for high-performance future computing systems. Compared to electronics, photonics has significant advantages for a fully parallel implementation of networks. A promising approach for parallel large-scale photonic networks is realizing the connections using diffraction. Here, we characterize the scalability of such diffractive coupling in great detail. Based on experiments, analytically obtained bounds and numerical simulations considering real-world optical imaging setups, we find that the concept in principle enables networks hosting over a million optical emitters. This would be a breakthrough in multiple areas, illustrating a clear path toward future, large scale photonic networks.
△ Less
Submitted 12 July, 2019;
originally announced July 2019.
-
Coupled nonlinear delay systems as deep convolutional neural networks
Authors:
Bogdan Penkovsky,
Xavier Porte,
Maxime Jacquot,
Laurent Larger,
Daniel Brunner
Abstract:
Neural networks are currently transforming the field of computer algorithms, yet their emulation on current computing substrates is highly inefficient. Reservoir computing was successfully implemented on a large variety of substrates and gave new insight in overcoming this implementation bottleneck. Despite its success, the approach lags behind the state of the art in deep learning. We therefore e…
▽ More
Neural networks are currently transforming the field of computer algorithms, yet their emulation on current computing substrates is highly inefficient. Reservoir computing was successfully implemented on a large variety of substrates and gave new insight in overcoming this implementation bottleneck. Despite its success, the approach lags behind the state of the art in deep learning. We therefore extend time-delay reservoirs to deep networks and demonstrate that these conceptually correspond to deep convolutional neural networks. Convolution is intrinsically realized on a substrate level by generic drive-response properties of dynamical systems. The resulting novelty is avoiding vector-matrix products between layers, which cause low efficiency in today's substrates. Compared to singleton time-delay reservoirs, our deep network achieves accuracy improvements by at least an order of magnitude in Mackey-Glass and Lorenz timeseries prediction.
△ Less
Submitted 5 July, 2019; v1 submitted 14 February, 2019;
originally announced February 2019.
-
Mutual coupling and synchronization of optically coupled quantum-dot micropillar lasers at ultra-low light levels
Authors:
Sören Kreinberg,
Xavier Porte,
David Schicke,
Benjamin Lingnau,
Christian Schneider,
Sven Höfling,
Kathy Lüdge,
Stephan Reitzenstein
Abstract:
In this work we explore the limits of synchronization of mutually coupled oscillators at the crossroads of classical and quantum physics. In order to address this uncovered regime of synchronization we apply electrically driven quantum dot micropillar lasers operating in the regime of cavity quantum electrodynamics. These high-$β$ microscale lasers feature cavity enhanced coupling of spontaneous e…
▽ More
In this work we explore the limits of synchronization of mutually coupled oscillators at the crossroads of classical and quantum physics. In order to address this uncovered regime of synchronization we apply electrically driven quantum dot micropillar lasers operating in the regime of cavity quantum electrodynamics. These high-$β$ microscale lasers feature cavity enhanced coupling of spontaneous emission and operate at output powers on the order of 100 nW. We selected pairs of micropillar lasers with almost identical optical properties in terms of the input-output dependence and the emission energy which we mutually couple over a distance of about 1m and bring into spectral resonance by precise temperature tuning. By excitation power and detuning dependent studies we unambiguously identify synchronization of two mutually coupled high-$β$ microlasers via frequency locking associated with a sub-GHz locking range. A detailed analysis of the synchronization behavior includes theoretical modeling based on semi-classical stochastic rate equations and reveals striking differences from optical synchronization in the classical domain with negligible spontaneous emission noise and optical powers usually well above the mW range. In particular, we observe deviations from the classically expected locking slope and broadened locking boundaries which are successfully explained by the fact the quantum noise plays an important role in our cavity enhanced optical oscillators. Beyond that, introducing additional self-feedback to the two mutually coupled microlasers allows us to realize zero-lag synchronization. Our work provides important insight into synchronization of optical oscillators at ultra-low light levels and has high potential to pave the way for future experiments in the quantum regime of synchronization.
△ Less
Submitted 4 August, 2018;
originally announced August 2018.
-
Quantum-optical spectroscopy of a two-level system using an electrically driven micropillar laser as resonant excitation source
Authors:
Sören Kreinberg,
Tomislav Grbešic,
Max Strauß,
Alexander Carmele,
Monika Emmerling,
Christian Schneider,
Sven Höfling,
Xavier Porte,
Stephan Reitzenstein
Abstract:
Two-level emitters constitute main building blocks of photonic quantum systems and are model systems for the exploration of quantum optics in the solid state. Most interesting is the strict-resonant excitation of such emitters to generate close to ideal quantum light and to control their occupation coherently. Up till now related experiments have been performed exclusively using bulky lasers which…
▽ More
Two-level emitters constitute main building blocks of photonic quantum systems and are model systems for the exploration of quantum optics in the solid state. Most interesting is the strict-resonant excitation of such emitters to generate close to ideal quantum light and to control their occupation coherently. Up till now related experiments have been performed exclusively using bulky lasers which hinders the application of resonantly driven two-level emitters in photonic quantum systems. Here we perform quantum-optical spectroscopy of a two-level system using a compact high-$β$ microlaser as excitation source. The two-level system is based on a semiconductor quantum dot (QD), which is excited resonantly by a fiber-coupled electrically driven micropillar laser. In this way we dress the excitonic state of the QD under continuous wave excitation and trigger the emission of single-photons with strong multi-photon suppression ($g^{(2)}(0)=0.02$) and high photon indistinguishably ($V=57\pm9\%$) via pulsed resonant excitation at 156 MHz.
△ Less
Submitted 27 November, 2017;
originally announced November 2017.
-
Micropillars with a controlled number of site-controlled quantum dots
Authors:
Arsenty Kaganskiy,
Fabian Gericke,
Tobias Heuser,
Tobias Heindel,
Xavier Porte,
Stephan Reitzenstein
Abstract:
We report on the realization of micropillars with site-controlled quantum dots (SCQDs) in the active layer. The SCQDs are grown via the buried stressor approach which allows for the positioned growth and device integration of a controllable number of QDs with high optical quality. This concept is very powerful as the number and the position of SCQDs in the cavity can be simultaneously controlled b…
▽ More
We report on the realization of micropillars with site-controlled quantum dots (SCQDs) in the active layer. The SCQDs are grown via the buried stressor approach which allows for the positioned growth and device integration of a controllable number of QDs with high optical quality. This concept is very powerful as the number and the position of SCQDs in the cavity can be simultaneously controlled by the design of the buried stressor. The fabricated micropillars exhibit a high degree of position control for the QDs above the buried stressor and $Q$-factors of up to 12000 at an emission wavelength around 930 nm. We experimentally analyze and numerically model the cavity $Q$-factor, the mode volume, the Purcell factor and the photon-extraction efficiency as a function of the aperture diameter of the buried stressor. Exploiting these SCQD micropillars, we experimentally observe the Purcell enhancement in the single-QD regime with $F_P$ = 4.3 $\pm$ 0.3.
△ Less
Submitted 25 November, 2017;
originally announced November 2017.
-
Controlling the gain contribution of background emitters in few-quantum-dot microlasers
Authors:
F. Gericke,
M. Segnon,
M. von Helversen,
C. Hopfmann,
T. Heindel,
C. Schneider,
S. Höfling,
M. Kamp,
A. Musiał,
X. Porte,
C. Gies,
S. Reitzenstein
Abstract:
We provide experimental and theoretical insight into single-emitter lasing effects in a quantum dot (QD)-microlaser under controlled variation of background gain provided by off-resonant discrete gain centers. For that purpose, we apply an advanced two-color excitation concept where the background gain contribution of off-resonant QDs can be continuously tuned by precisely balancing the relative e…
▽ More
We provide experimental and theoretical insight into single-emitter lasing effects in a quantum dot (QD)-microlaser under controlled variation of background gain provided by off-resonant discrete gain centers. For that purpose, we apply an advanced two-color excitation concept where the background gain contribution of off-resonant QDs can be continuously tuned by precisely balancing the relative excitation power of two lasers emitting at different wavelengths. In this way, by selectively exciting a single resonant QD and off-resonant QDs, we identify distinct single-QD signatures in the lasing characteristics and distinguish between gain contributions of a single resonant emitter and a countable number of off-resonant background emitters to the optical output of the microlaser. We address the important question whether single-QD lasing is feasible in experimentally accessible systems and show that, for the investigated microlaser, the single-QD gain needs to be supported by the background gain contribution of off-resonant QDs to reach the transition to lasing. Interestingly, while a single QD cannot drive the investigated micropillar into lasing, its relative contribution to the emission can be as high as 70% and it dominates the statistics of emitted photons in the intermediate excitation regime below threshold.
△ Less
Submitted 12 April, 2017;
originally announced April 2017.
