arXiv subjects

Michał Matuszewski

Publications and source records attributed to Michał Matuszewski.

At least 19 recordsLinked to original sources

arXiv2026

Near-Equilibrium Propagation training in nonlinear wave systems

Backpropagation learning algorithm, the workhorse of modern artificial intelligence, is notoriously difficult to implement in physical neural networks. Equilibrium Propagation (EP) is an alternative with comparable efficiency and strong potential for in-situ training. We extend EP learning to both discrete and continuous complex-valued wave systems. In contrast to previous EP implementations, our scheme is valid in the weakly dissipative regime, and readily applicable to a wide range of physical settings, even without well defined nodes, where trainable inter-node connections can be replaced by trainable local potential. We test the method in driven-dissipative exciton-polariton condensates governed by generalized Gross-Pitaevskii dynamics. Numerical studies on standard benchmarks, including a simple logical task and handwritten-digit recognition, demonstrate stable convergence, establishing a practical route to in-situ learning in physical systems in which system control is restricted to local parameters.

cs.LG↗

arXiv2026

Quantization of Polaritons Confined in Dielectric Structures

Light-matter interaction in the regime of strong quantum coupling is usually treated within the framework of the Hopfield model. However, the picture of coupling well-defined modes of light and matter is correct only as long as the shapes of these eigenmodes are not substantially modified by the interaction. Moreover, parameters of theoretical models are usually obtained by fitting to experimental data. To date, there is no straightforward method to determine a quantum master equation corresponding to a system with specific dielectric structure, which may lead to incompatibility of theoretical descriptions and physical realizations. In this work, a recipe for obtaining a quantum model in the polariton eigenmode basis is presented, based on Bogoliubov transformation in the conservative case and third quantization technique in the dissipative case. It is shown how this method can be used for boosting interaction strength and engineering nonlocal many-body interactions in carefully designed nanostructures, resulting in strongly nonclassical correlations of emitted light.

quant-ph↗

arXiv2026

Antibunching in locally driven dissipative Lieb lattices

In Lieb lattices, geometric frustration and destructive interference of hopping cancels the occupation of certain sites, leading to flat-band physics. Here, we show numerically how, in the driven-dissipative Bose-Hubbard (DDBH) model arranged into Lieb lattices and related geometries, specific localised driving schemes can repurpose this interference to generate enhanced antibunching via a mechanism similar to the so-called unconventional photon blockade. Stochastic simulations using the positive-P method allow us to calculate occupations and second order correlations exactly for extended lattices. We use this to optimise the parameters for the possible observation of this effect in polariton micropillar experiments. This work demonstrates the possibility of using localised driving and interference effects to generate non-trivial quantum correlations in open quantum lattice systems. Specifically, producing antibunching in the dark sites of the flat band system rather than the usual and less useful bunching.

quant-ph↗

arXiv2026

Strongly Nonlinear Slow Light Polaritons in Subwavelength Modulated Waveguides

Slow light is a regime of reduced group velocity, resulting in increased photon density in optical pulses and enhanced nonlinear effects. Here, we propose the realization of slow light in the regime of strong light-matter interaction between waveguide photons and semiconductor excitons. We design a dielectric superlattice structure with a nearly-flat band characterized by low group velocity and group velocity dispersion, both required for enhancing nonlinear effects with ultrashort pulses. Furthermore, by applying this general framework to a perovskite-based structure, we demonstrate an enhancement of the single-particle phase shift by a factor of more than 20, representing a significant step toward the few-photon quantum regime. Our results provide a blueprint for accessible strong interactions in solid-state integrated optics.

physics.optics↗

arXiv2026

Quantum Light Detection with Enhanced Photonic Neural Network

Advances in quantum technologies are accelerating the demand for optical quantum state sensors that combine high precision, versatility, and scalability within a unified hardware platform. Quantum reservoir computing offers a powerful route toward this goal by exploiting the nonlinear dynamics of quantum systems to process and interpret quantum information efficiently. Photonic neural networks are particularly well suited for such implementations, owing to their intrinsic sensitivity to photon-encoded quantum information. However, the practical realisation of photonic quantum reservoirs remains constrained by the inherently weak optical nonlinearities of available materials and the technological challenges of fabricating densely coupled quantum networks. To address these limitations, we introduce a hybrid quantum-classical detection protocol that integrates the advantages of quantum reservoirs with the adaptive learning capabilities of analogue neural networks. This synergistic architecture substantially enhances information-extraction accuracy and robustness, enabling low-cost performance improvements of quantum light sensors. Based on the proposed approach, we achieved significant improvements in quantum state classification, tomography, and feature regression, even for reservoirs with a relatively small nonlinearity-to-losses ratio $U/γ\approx 0.02$ in a network of only five nodes. By reducing reliance on material nonlinearity and reservoir size, the proposed approach facilitates the practical deployment of high-fidelity photonic quantum sensors on existing integrated platforms, paving the way toward chip-scale quantum processors and photonic sensing technologies.

quant-ph↗

arXiv2025

Spectroscopy on a single nonlinear mode recognizes quantum states

Characterising optical quantum states is essential for the development of quantum technologies. While traditional approaches to perform full quantum state tomography are often experimentally demanding, neuromorphic architectures may provide an effective alternative. In this work, we demonstrate how a quantum nonlinear driven-dissipative mode is sufficient to act as a quantum reservoir. By analyzing the occupations at different frequencies in the emission spectrum, a linear regression suffices in many cases to recognize the relevant parameters of incident squeezed states. Beyond highlighting the general potential of this approach under continuous driving, we illustrate its effectiveness in an explicit nontrivial example where the source is a degenerate optical parametric oscillator (OPO), coupled to a nonlinear polariton microcavity.

quant-ph↗

arXiv2025

Phase-Space Framework for Noisy Intermediate-Scale Quantum Optical Neural Networks

Quantum optical neural networks (QONNs) enable information processing beyond classical limits by exploiting the advantages of classical and quantum optics. However, simulation of large-scale bosonic lattices remains a significant challenge due to the exponential growth of the Hilbert space required to describe a quantum network accurately. Consequently, previous theoretical studies have been limited to small-scale systems, leaving the behaviour of multimode QONNs largely unexplored. This work presents an efficient computational framework based on the phase-space positive-P method for simulating bosonic neuromorphic systems. This approach provides a view to previously inaccessible regimes, allowing the validation of large-scale bosonic networks in various quantum machine learning tasks such as quantum state classification and quantum state feature prediction. Our results show that the performance of a large quantum reservoir does not improve monotonously with the number of bosonic modes, instead following a complex dependence driven by the interplay of nonlinearity, reservoir size, and the average occupation of the input mode. These findings are essential for designing and optimising optical bosonic reservoirs for future quantum neuromorphic computing devices.

quant-ph↗

arXiv2025

Modeling Nonlinear Optics with the Transfer Matrix Method

The Transfer Matrix Method (TMM) is a widely used technique for modeling linear propagation of electromagnetic waves through stratified layered media. However, since its extension to inhomogeneous and nonlinear systems is not straightforward, much more computationally demanding methods such as Finite-difference time-domain (FDTD) or Method of lines (MoL) are typically used. In this work, we extend the TMM framework to incorporate the effects of nonlinearity. We consider the case when strong coupling between excitons (electron-hole pairs) and photons leads to the formation of exciton-polaritons. This extension is crucial for accurately simulating the behavior of light in polariton microcavities, where nonlinearities arising from exciton-exciton interactions play a key role. We perform efficient simulations of light transmission and reflection in a multidimensional system using the plane wave basis. Additionally, we compare our extended TMM approach with the state-of-the-art admittance transfer method, and highlight the computational advantage of extended TMM for large-scale systems. The extended TMM not only provides a robust and computationally efficient numerical framework, but also paves the way for the development of future low-power nonlinear optical devices, polariton-based photonic circuits, and quantum photonic technologies.

physics.optics↗

arXiv2025

Estimation of the second-order coherence function using quantum reservoir and ensemble methods

We propose a machine learning-based approach enhanced by quantum reservoir computing (QRC) to estimate the zero-time second-order correlation function g2(0). Typically, measuring g2(0) requires single-photon detectors and time-correlated measurements. Machine learning may offer practical solutions by training a model to estimate g2(0) solely from average intensity measurements. In our method, emission from a given quantum source is first processed in QRC. During the inference phase, only intensity measurements are used, which are then passed to a software-based decision tree-based ensemble model. We evaluate this hybrid quantum-classical approach across a variety of quantum optical systems and demonstrate that it provides accurate estimates of g2(0). We further extend our analysis to assess the ability of a trained model to generalize beyond its training distribution, both to the same system under different physical parameters and to fundamentally different quantum sources. While the model may yield reliable estimates within specific regimes, its performance across distinct systems is generally limited.

quant-ph↗

arXiv2024

Room temperature exciton-polariton neural network with perovskite crystal

Limitations of electronics have stimulated the search for novel unconventional computing platforms that enable energy-efficient and ultra-fast information processing. Among various systems, exciton-polaritons stand out as promising candidates for the realization of optical neuromorphic devices. This is due to their unique hybrid light-matter properties, resulting in strong optical nonlinearity and excellent transport capabilities. However, previous implementations of polariton neural networks have been restricted to cryogenic temperatures, limiting their practical applications. In this work, using non-equillibrium Bose-Einstein condensation in a monocrystalline perovskite waveguide, we demonstrate the first room-temperature exciton-polariton neural network. Its performance is verified in various machine learning tasks, including binary classification, and object detection. Our result is a crucial milestone in the development of practical applications of polariton neural networks and provides new perspectives for optical computing accelerators based on perovskites.

physics.optics↗

arXiv2024

Non-Hermitian polariton-photon coupling in a perovskite open microcavity

Exploring the non-Hermitian properties of semiconductor materials for optical applications is at the forefront of photonic research. However, the selection of appropriate systems to implement such photonic devices remains a topic of debate. In this work, we demonstrate that a perovskite crystal, characterized by its easy and low-cost manufacturing, when placed between two distributed Bragg reflectors with an air gap, can form a natural double microcavity. This construction shows promising properties for the realisation of novel, tunable non-Hermitian photonic devices through strong light-matter coupling. We reveal that such a system exhibits double-coupled polariton modes with dispersion including multiple inflection points. Owing to its non-Hermiticity, our system exhibits nonreciprocal properties and allows for the observation of exceptional points. Our experimental studies are in agreement with the theoretical analysis based on coupled mode theory and calculations based on transfer matrix method.

cond-mat.mes-hall↗

arXiv2024

Analogue Spin Simulators: How to keep the Amplitude Homogeneous

A setup that simulates ground states of spin graphs would allow one to solve computationally hard optimisation problems efficiently. Current optical setups to this goal have difficulties decoupling the amplitude and phase degrees of freedom of each effective spin; risking to yield the mapping invalid, a problem known as amplitude heterogeneity. Here, we propose a setup with coupled active optical cavity modes, where this problem is eliminated through their particular geometric arrangement. Acting as an effective Monte Carlo solver, the ground state can be found exactly. By tuning a parameter, the setup solves XY or Ising problems.

quant-ph↗

arXiv2023

Harnessing exciton-polaritons for digital computing, neuromorphic computing, and optimization

Polaritons are quasiparticles resulting from strong quantum coupling of light and matter. Peculiar properties of polaritons are a mixture of physics usually restricted to one of these realms, making them interesting for study not only from the fundamental point of view but also for applications. In recent years, many studies have been devoted to the potential use of exciton-polaritons for computing. Very recently, it has been shown experimentally that they can be harnessed not only for digital computing, but also for optical neural networks and for optimization related to hard computational problems. Here, we provide a brief review of recent studies and most important results in this area. We focus our attention in particular on the emerging concepts of non-von-Neumann computing schemes and their realizations in exciton-polariton systems.

cond-mat.quant-gas↗

arXiv2023

Exceptional points and phase transitions in non-Hermitian binary systems

Recent study demonstrated that steady states of a polariton system may show a first-order dissipative phase transition with an exceptional point that appears as an endpoint of the phase boundary [R. Hanai et al., Phys. Rev. Lett. 122, 185301 (2019)]. Here, we show that this phase transition is strictly related to the stability of solutions. In general, the exceptional point does not correspond to the endpoint of a phase transition, but rather it is the point where stable and unstable solutions coalesce. Moreover, we show that the transition may occur also in the weak coupling regime, which was excluded previously. In a certain range of parameters, we demonstrate permanent Rabi-like oscillations between light and matter fields. Our results contribute to the understanding of nonequilibrium light-matter systems, but can be generalized to any two-component oscillatory systems with gain and loss.

quant-ph↗

arXiv2023

The role of all-optical neural networks

In light of recent achievements in optical computing and machine learning, we consider the conditions under which all-optical computing may surpass electronic and optoelectronic computing in terms of energy efficiency and scalability. When considering the performance of a system as a whole, the cost of memory access and data acquisition is likely to be one of the main efficiency bottlenecks not only for electronic, but also for optoelectronic and all-optical devices. However, we predict that all-optical devices will be at an advantage in the case of inference in large neural network models, and the advantage will be particularly large in the case of generative models. We also consider the limitations of all-optical neural networks including footprint, strength of nonlinearity, optical signal degradation, limited precision of computations, and quantum noise.

cs.ET↗

arXiv2023

Natural exceptional points in the excitation spectrum of a light-matter system

We observe natural exceptional points in the excitation spectrum of an exciton-polariton system by optically tuning the light-matter interactions. The observed exceptional points do not require any spatial or polarization degrees of freedom and result solely from the transition from weak to strong light-matter coupling. We demonstrate that they do not coincide with the threshold for photon lasing, confirming previous theoretical predictions [Phys. Rev. Lett. 122, 185301 (2019), Optica 7, 1015 (2020) ]. Using a technique where a strong coherent laser pump induces up-converted excitations, we encircle the exceptional point in the parameter space of coupling strength and particle momentum. Our method of local optical control of light-matter coupling paves the way to investigation of fundamental phenomena including dissipative phase transitions and non-Hermitian topological states.

physics.optics↗

arXiv2023

Non-Hermitian synthetic lattices with light-matter coupling

We propose that light-matter coupling can be used to realize synthetic lattices. In particular, we consider a one-dimensional chain of exciton-photon sites to create a comb lattice that exhibits a transition from a flat band to a finite mass dispersion by tuning site-dependent light-matter coupling. Moreover, in a non-Hermitian system with gain and loss, the flat band phase is much more robust and the transition is accompanied by the appearance of exceptional points in the complex energy spectrum. We demonstrate that by engineering the light-matter coupling in the synthetic lattice, one can explore various phases in the lasing regime. Our proposal paves the way for studying non-Hermitian systems in higher dimensions.

cond-mat.mes-hall↗

arXiv2022

Training a neural network with exciton-polariton optical nonlinearity

In contrast to software simulations of neural networks, hardware implementations have often limited or no tunability. While such networks promise great improvements in terms of speed and energy efficiency, their performance is limited by the difficulty to apply efficient training. We propose and realize experimentally an optical system where highly efficient backpropagation training can be applied through an array of highly nonlinear, non-tunable nodes. The system includes exciton-polariton nodes realizing nonlinear activation functions. We demonstrate a high classification accuracy in the MNIST handwritten digit benchmark in a single hidden layer system.

cs.LG↗