-
Light-cone feature selection in methane hyperspectral images
Authors:
Artur Miroszewski,
Jakub Nalepa,
Agata M. Wijata
Abstract:
Hyperspectral images (HSIs) capture detailed spectral information across numerous contiguous bands, enabling the extraction of intrinsic characteristics of scanned objects and areas. This study focuses on the application of light-cone feature selection in quantum machine learning for methane detection and localization using HSIs. The proposed method leverages quantum methods to enhance feature sel…
▽ More
Hyperspectral images (HSIs) capture detailed spectral information across numerous contiguous bands, enabling the extraction of intrinsic characteristics of scanned objects and areas. This study focuses on the application of light-cone feature selection in quantum machine learning for methane detection and localization using HSIs. The proposed method leverages quantum methods to enhance feature selection and classification accuracy. The dataset used includes HSIs collected by the AVIRIS-NG instrument captured in geographically diverse locations. In this study, we investigate the performance of support vector machine classifiers with different classic and quantum kernels. The results indicate that the quantum kernel classifier, combined with light-cone feature selection, provides in one metric, superior performance when compared to the classic techniques. It demonstrates the potential of quantum machine learning in improving the remote sensing data analysis for environmental monitoring.
△ Less
Submitted 1 April, 2025;
originally announced April 2025.
-
To center or not to center? Hyperspectral data vs. quantum covariance matrices
Authors:
Artur Miroszewski,
Jakub Nalepa,
Agata M. Wijata
Abstract:
We highlight how the $L^2$ normalization required for embedding data in quantum states affects data centering, which can significantly influence quantum amplitude-encoded covariance matrices in quantum data analysis algorithms. We examine the spectra and eigenvectors of quantum covariance matrices derived from hyperspectral data under various centering scenarios. Surprisingly, our findings reveal…
▽ More
We highlight how the $L^2$ normalization required for embedding data in quantum states affects data centering, which can significantly influence quantum amplitude-encoded covariance matrices in quantum data analysis algorithms. We examine the spectra and eigenvectors of quantum covariance matrices derived from hyperspectral data under various centering scenarios. Surprisingly, our findings reveal that classification performance in problems reduced by principal component analysis remains unaffected, no matter if the data is centered or uncentered, provided that eigenvector filtering is handled appropriately.
△ Less
Submitted 1 April, 2025;
originally announced April 2025.
-
In Search of Quantum Advantage: Estimating the Number of Shots in Quantum Kernel Methods
Authors:
Artur Miroszewski,
Marco Fellous Asiani,
Jakub Mielczarek,
Bertrand Le Saux,
Jakub Nalepa
Abstract:
Quantum Machine Learning (QML) has gathered significant attention through approaches like Quantum Kernel Machines. While these methods hold considerable promise, their quantum nature presents inherent challenges. One major challenge is the limited resolution of estimated kernel values caused by the finite number of circuit runs performed on a quantum device. In this study, we propose a comprehensi…
▽ More
Quantum Machine Learning (QML) has gathered significant attention through approaches like Quantum Kernel Machines. While these methods hold considerable promise, their quantum nature presents inherent challenges. One major challenge is the limited resolution of estimated kernel values caused by the finite number of circuit runs performed on a quantum device. In this study, we propose a comprehensive system of rules and heuristics for estimating the required number of circuit runs in quantum kernel methods. We introduce two critical effects that necessitate an increased measurement precision through additional circuit runs: the spread effect and the concentration effect. The effects are analyzed in the context of fidelity and projected quantum kernels. To address these phenomena, we develop an approach for estimating desired precision of kernel values, which, in turn, is translated into the number of circuit runs. Our methodology is validated through extensive numerical simulations, focusing on the problem of exponential value concentration. We stress that quantum kernel methods should not only be considered from the machine learning performance perspective, but also from the context of the resource consumption. The results provide insights into the possible benefits of quantum kernel methods, offering a guidance for their application in quantum machine learning tasks.
△ Less
Submitted 22 July, 2024;
originally announced July 2024.
-
Quantum Machine Learning for Remote Sensing: Exploring potential and challenges
Authors:
Artur Miroszewski,
Jakub Nalepa,
Bertrand Le Saux,
Jakub Mielczarek
Abstract:
The industry of quantum technologies is rapidly expanding, offering promising opportunities for various scientific domains. Among these emerging technologies, Quantum Machine Learning (QML) has attracted considerable attention due to its potential to revolutionize data processing and analysis. In this paper, we investigate the application of QML in the field of remote sensing. It is believed that…
▽ More
The industry of quantum technologies is rapidly expanding, offering promising opportunities for various scientific domains. Among these emerging technologies, Quantum Machine Learning (QML) has attracted considerable attention due to its potential to revolutionize data processing and analysis. In this paper, we investigate the application of QML in the field of remote sensing. It is believed that QML can provide valuable insights for analysis of data from space. We delve into the common beliefs surrounding the quantum advantage in QML for remote sensing and highlight the open challenges that need to be addressed. To shed light on the challenges, we conduct a study focused on the problem of kernel value concentration, a phenomenon that adversely affects the runtime of quantum computers. Our findings indicate that while this issue negatively impacts quantum computer performance, it does not entirely negate the potential quantum advantage in QML for remote sensing.
△ Less
Submitted 13 November, 2023;
originally announced November 2023.
-
Cloud Detection in Multispectral Satellite Images Using Support Vector Machines With Quantum Kernels
Authors:
Artur Miroszewski,
Jakub Mielczarek,
Filip Szczepanek,
Grzegorz Czelusta,
Bartosz Grabowski,
Bertrand Le Saux,
Jakub Nalepa
Abstract:
Support vector machines (SVMs) are a well-established classifier effectively deployed in an array of pattern recognition and classification tasks. In this work, we consider extending classic SVMs with quantum kernels and applying them to satellite data analysis. The design and implementation of SVMs with quantum kernels (hybrid SVMs) is presented. It consists of the Quantum Kernel Estimation (QKE)…
▽ More
Support vector machines (SVMs) are a well-established classifier effectively deployed in an array of pattern recognition and classification tasks. In this work, we consider extending classic SVMs with quantum kernels and applying them to satellite data analysis. The design and implementation of SVMs with quantum kernels (hybrid SVMs) is presented. It consists of the Quantum Kernel Estimation (QKE) procedure combined with a classic SVM training routine. The pixel data are mapped to the Hilbert space using ZZ-feature maps acting on the parameterized ansatz state. The parameters are optimized to maximize the kernel target alignment. We approach the problem of cloud detection in satellite image data, which is one of the pivotal steps in both on-the-ground and on-board satellite image analysis processing chains. The experiments performed over the benchmark Landsat-8 multispectral dataset revealed that the simulated hybrid SVM successfully classifies satellite images with accuracy on par with classic SVMs.
△ Less
Submitted 14 July, 2023;
originally announced July 2023.
-
Optimizing Kernel-Target Alignment for cloud detection in multispectral satellite images
Authors:
Artur Miroszewski,
Jakub Mielczarek,
Filip Szczepanek,
Grzegorz Czelusta,
Bartosz Grabowski,
Bertrand Le Saux,
Jakub Nalepa
Abstract:
The optimization of Kernel-Target Alignment (TA) has been recently proposed as a way to reduce the number of hardware resources in quantum classifiers. It allows to exchange highly expressive and costly circuits to moderate size, task oriented ones. In this work we propose a simple toy model to study the optimization landscape of the Kernel-Target Alignment. We find that for underparameterized cir…
▽ More
The optimization of Kernel-Target Alignment (TA) has been recently proposed as a way to reduce the number of hardware resources in quantum classifiers. It allows to exchange highly expressive and costly circuits to moderate size, task oriented ones. In this work we propose a simple toy model to study the optimization landscape of the Kernel-Target Alignment. We find that for underparameterized circuits the optimization landscape possess either many local extrema or becomes flat with narrow global extremum. We find the dependence of the width of the global extremum peak on the amount of data introduced to the model. The experimental study was performed using multispectral satellite data, and we targeted the cloud detection task, being one of the most fundamental and important image analysis tasks in remote sensing.
△ Less
Submitted 26 June, 2023;
originally announced June 2023.
-
Detecting Clouds in Multispectral Satellite Images Using Quantum-Kernel Support Vector Machines
Authors:
Artur Miroszewski,
Jakub Mielczarek,
Grzegorz Czelusta,
Filip Szczepanek,
Bartosz Grabowski,
Bertrand Le Saux,
Jakub Nalepa
Abstract:
Support vector machines (SVMs) are a well-established classifier effectively deployed in an array of classification tasks. In this work, we consider extending classical SVMs with quantum kernels and applying them to satellite data analysis. The design and implementation of SVMs with quantum kernels (hybrid SVMs) are presented. Here, the pixels are mapped to the Hilbert space using a family of para…
▽ More
Support vector machines (SVMs) are a well-established classifier effectively deployed in an array of classification tasks. In this work, we consider extending classical SVMs with quantum kernels and applying them to satellite data analysis. The design and implementation of SVMs with quantum kernels (hybrid SVMs) are presented. Here, the pixels are mapped to the Hilbert space using a family of parameterized quantum feature maps (related to quantum kernels). The parameters are optimized to maximize the kernel target alignment. The quantum kernels have been selected such that they enabled analysis of numerous relevant properties while being able to simulate them with classical computers on a real-life large-scale dataset. Specifically, we approach the problem of cloud detection in the multispectral satellite imagery, which is one of the pivotal steps in both on-the-ground and on-board satellite image analysis processing chains. The experiments performed over the benchmark Landsat-8 multispectral dataset revealed that the simulated hybrid SVM successfully classifies satellite images with accuracy comparable to the classical SVM with the RBF kernel for large datasets. Interestingly, for large datasets, the high accuracy was also observed for the simple quantum kernels, lacking quantum entanglement.
△ Less
Submitted 16 February, 2023;
originally announced February 2023.
-
Dynamics of primordial fields in quantum cosmological spacetimes
Authors:
Przemysław Małkiewicz,
Artur Miroszewski
Abstract:
Quantum cosmological models are commonly described by means of semiclassical approximations in which a smooth evolution of the expectation values of elementary geometry operators replaces the classical and singular dynamics. The advantage of such descriptions is that they are relatively simple and display the classical behavior for large universes. However, they may smooth out an important inner s…
▽ More
Quantum cosmological models are commonly described by means of semiclassical approximations in which a smooth evolution of the expectation values of elementary geometry operators replaces the classical and singular dynamics. The advantage of such descriptions is that they are relatively simple and display the classical behavior for large universes. However, they may smooth out an important inner structure and to include it a more detailed treatment is needed. The purpose of the present work is to investigate quantum uncertainty in the basic background variables and its influence on primordial gravitational waves. To this end we quantize a model of the Friedmann-Lemaitre-Robertson-Walker universe filled with a linear barotropic cosmological fluid and with gravitational waves. We carefully derive the dynamical equations for the perturbations in quantum spacetime. The quantization yields an equation of motion for the Fourier modes of gravitational radiation, which is a quantum extension to the usual parametric oscillator equation for gravitational waves propagating in an expanding universe. The two quantum effects from the cosmological background that enter the enhanced equation of motion are (i) a repulsive potential resolving the big bang singularity and replacing it with a big bounce and (ii) uncertainties in the numerical values for the background spacetime dynamical variables. First we study the former effect and its consequences for the primordial amplitude spectrum and carefully discuss the relation between the bounce scale and the physical predictions of the model. Next we investigate the latter effect, in particular the extent to which it may affect the primordial amplitude of gravitational waves. Making use of the WKB approximation we find an analytical formula for the amplitude spectrum as a function of the quantum dispersion of the background spacetime.
△ Less
Submitted 27 April, 2021; v1 submitted 6 November, 2020;
originally announced November 2020.
-
Quantum dynamics in Weyl-Heisenberg coherent states
Authors:
Artur Miroszewski
Abstract:
The article explores a new formalism for describing motion in quantum mechanics. The construction is based on generalized coherent states with evolving fiducial vector. Weyl-Heisenberg coherent states are utilised to split quantum systems into `classical' and `quantum' degrees of freedom. The decomposition is found to be equivalent to quantum mechanics perceived from a semi-classical frame. The sp…
▽ More
The article explores a new formalism for describing motion in quantum mechanics. The construction is based on generalized coherent states with evolving fiducial vector. Weyl-Heisenberg coherent states are utilised to split quantum systems into `classical' and `quantum' degrees of freedom. The decomposition is found to be equivalent to quantum mechanics perceived from a semi-classical frame. The split allows for introduction of a new definition of classical state and is a convenient starting point for approximate analysis of quantum dynamics. An example of a meta-stable state is given as a practical illustration of the introduced concepts.
△ Less
Submitted 9 September, 2020; v1 submitted 31 August, 2020;
originally announced September 2020.
-
Quantum phase space trajectories with application to quantum cosmology
Authors:
Przemysław Małkiewicz,
Artur Miroszewski,
Hervé Bergeron
Abstract:
We develop an approach to quantum dynamics based on quantum phase space trajectories. The latter are built from a unitary irreducible representation of the symmetry group of the respective classical phase space. We use a quantum action functional to derive the basic equations. In principle, our formulation is equivalent to the Hilbert space formulation. However, the former allows for consistent tr…
▽ More
We develop an approach to quantum dynamics based on quantum phase space trajectories. The latter are built from a unitary irreducible representation of the symmetry group of the respective classical phase space. We use a quantum action functional to derive the basic equations. In principle, our formulation is equivalent to the Hilbert space formulation. However, the former allows for consistent truncations to reduced phase spaces in which approximate quantum dynamics can be derived. We believe that our approach can be very useful in the domain of quantum cosmology and therefore, we use the cosmological phase space example to establish the basic equations of this formalism.
△ Less
Submitted 5 March, 2018; v1 submitted 13 December, 2017;
originally announced December 2017.
-
Time crystal platform: from quasi-crystal structures in time to systems with exotic interactions
Authors:
Krzysztof Giergiel,
Artur Miroszewski,
Krzysztof Sacha
Abstract:
Time crystals are quantum many-body systems which, due to interactions between particles, are able to spontaneously self-organize their motion in a periodic way in time by analogy with the formation of crystalline structures in space in condensed matter physics. In solid state physics properties of space crystals are often investigated with the help of external potentials that are spatially period…
▽ More
Time crystals are quantum many-body systems which, due to interactions between particles, are able to spontaneously self-organize their motion in a periodic way in time by analogy with the formation of crystalline structures in space in condensed matter physics. In solid state physics properties of space crystals are often investigated with the help of external potentials that are spatially periodic and reflect various crystalline structures. A similar approach can be applied for time crystals, as periodically driven systems constitute counterparts of spatially periodic systems, but in the time domain. Here we show that condensed matter problems ranging from single particles in potentials of quasi-crystal structure to many-body systems with exotic long-range interactions can be realized in the time domain with an appropriate periodic driving. Moreover, it is possible to create molecules where atoms are bound together due to destructive interference if the atomic scattering length is modulated in time.
△ Less
Submitted 3 April, 2018; v1 submitted 27 October, 2017;
originally announced October 2017.
-
Internal clock formulation of quantum mechanics
Authors:
Przemysław Małkiewicz,
Artur Miroszewski
Abstract:
The basic tenet of the present work is the assumption of the lack of external and fixed time in the Universe. This assumption is best embodied by general relativity, which replaces the fixed space-time structure with the gravitational field, which is subject to dynamics. The lack of time does not imply the lack of evolution but rather brings to the forefront the role of internal clocks which are s…
▽ More
The basic tenet of the present work is the assumption of the lack of external and fixed time in the Universe. This assumption is best embodied by general relativity, which replaces the fixed space-time structure with the gravitational field, which is subject to dynamics. The lack of time does not imply the lack of evolution but rather brings to the forefront the role of internal clocks which are some largely arbitrary internal degrees of freedom with respect to which the evolution of timeless systems can be described. We take this idea seriously and try to understand what it implies for quantum mechanics when the fixed external time is replaced by an arbitrary internal clock. We put the issue in a solid, mathematically rigorous framework. We find that the dynamical interpretation of a quantum state of a timeless system depends on the employed internal clock. In particular, we find that the continuous spectra of well-known dynamical observables like the position of a free particle on the real line may turn discrete if measured in unusual clocks. We discuss the meaning of our result for attempts at quantization of global gravitational degrees of freedom.
△ Less
Submitted 18 September, 2017; v1 submitted 2 June, 2017;
originally announced June 2017.
