SEARCH · Searcharxiv

Results for “cond-mat.dis-nn”

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 235 records · Page 13Linked to original sources

arXiv2017

Pre-freezing transition in Boltzmann-Gibbs measures associated with log-correlated fields

We consider Boltzmann-Gibbs measures associated with log-correlated Gaussian fields as potentials and study their multifractal properties which exhibit phase transitions. In particular, the pre-freezing and freezing phenomena of the annealed exponent, predicted by Fyodorov using a modified replica-symmetry-breaking ansatz, are generalised to arbitrary dimension and verified using results from Gaussian multiplicative chaos theory.

math.PR↗

arXiv2018

State equation from the spectral structure of human brain activity

Neural electromagnetic (EM) signals recorded non-invasively from individual human subjects vary in complexity and magnitude. Nonetheless, variation in neural activity has been difficult to quantify and interpret, due to complex, broad-band features in the frequency domain. Studying signals recorded with magnetoencephalography (MEG) from healthy young adult subjects while in resting and active states, a systematic framework inspired by thermodynamics is applied to neural EM signals. Despite considerable inter-subject variation in terms of spectral entropy and energy across time epochs, data support the existence of a robust and linear relationship defining an effective state equation, with higher energy and lower entropy in the resting state compared to active, consistently across subjects. Mechanisms underlying the emergence of relationships between empirically measured effective state functions are further investigated using a model network of coupled oscillators, suggesting an interplay between noise and coupling strength can account for coherent variation of empirically observed quantities. Taken together, the results show macroscopic neural observables follow a robust, non-trivial conservation rule for energy modulation and information generation.

q-bio.NC↗

arXiv2018

Optimization of neural networks via finite-value quantum fluctuations

We numerically test an optimization method for deep neural networks (DNNs) using quantum fluctuations inspired by quantum annealing. For efficient optimization, our method utilizes the quantum tunneling effect beyond the potential barriers. The path integral formulation of the DNN optimization generates an attracting force to simulate the quantum tunneling effect. In the standard quantum annealing method, the quantum fluctuations will vanish at the last stage of optimization. In this study, we propose a learning protocol that utilizes a finite value for quantum fluctuations strength to obtain higher generalization performance, which is a type of robustness. We demonstrate the performance of our method using two well-known open datasets: the MNIST dataset and the Olivetti face dataset. Although computational costs prevent us from testing our method on large datasets with high-dimensional data, results show that our method can enhance generalization performance by induction of the finite value for quantum fluctuations.

cond-mat.dis-nn↗

arXiv2018

Role of thermal expansion heterogeneity in the cryogenic rejuvenation of metallic glasses

Cryogenic rejuvenation in metallic glasses reported in Ketov et al 's experiment (Nature(2015)524,200) has attracted much attention, both in experiments and numerical studies. The atomic mechanism of rejuvenation has been conjectured to be related to the heterogeneity of the glassy state, but the quantitative evidence is still elusive. Here we use molecular dynamics simulations of a model metallic glass to investigate the heterogeneity in the local thermal expansion. We then combine the resulting spatial distribution of thermal expansion with a continuum mechanics calculation to infer the internal stresses caused by a thermal cycle. Comparing the internal stress with the local yield stress, we prove that the heterogeneity in thermomechanical response has the potential to trigger local shear transformations, and therefore to induce rejuvenation during a cryogenic thermal cycling.

cond-mat.dis-nn↗

arXiv2018

Observation of Bulk Polarization Transitions and Higher-Order Embedded Topological Eigenstates for Sound

Topological systems are inherently robust to disorder and continuous perturbations, resulting in dissipation-free edge transport of electrons in quantum solids, or reflectionless guiding of photons and phonons in classical wave systems characterized by Chern or spin-Chern topological invariants. These established examples of topological physics, however, do not exhaust all possible topological phases, and recently a new class of topological metamaterials characterized by bulk polarization has been introduced. In addition to edge conduction, these systems have been shown to host higher-order topological modes. Here, we introduce and measure topological bulk polarization in 3D printed two-dimensional acoustic meta-structures, and observe topological transitions as the design parameters are tuned. We also demonstrate that our topological meta-structure hosts both 1D edge and higher-order 0D corner states with unique acoustic properties. The edge states have spin polarization that reverses for opposite propagation direction, thus supporting directional excitation. Corner states are pinned to the meta-structure corners, and rapidly decay both along the edges and into the bulk. Interestingly, these 0D states can spectrally overlap with the continuum of bulk states, but are not compatible with radiation, thus enabling embedded topological eigenstates within the continuum of bulk modes. Their confinement and inherent topological robustness is experimentally confirmed by deliberately introducing disorder. Our findings open new directions in acoustics for advanced sound propagation and manipulation.

cond-mat.mtrl-sci↗

arXiv2018

Characterizing the complexity of time series network graphs: A simplicial approach

We analyze the time series obtained from different dynamical regimes of the logistic map by constructing their equivalent time series (TS) networks, using the visibility algorithm. The regimes analyzed include both periodic and chaotic regimes, as well as intermittent regimes and the Feigenbaum attractor at the edge of chaos. We use the methods of algebraic topology to define the simplicial characterizers, which can analyse the simplicial structure of the networks at both the global and local levels. The simplicial characterisers bring out the hierarchical levels of complexity at various topological levels. These hierarchical levels of complexity find the skeleton of the local dynamics embedded in the network which influence the global dynamical properties of the system, and also permit the identification of dominant motifs. We also analyze the same networks using conventional network characterizers such as average path lengths and clustering coefficients. We see that the simplicial characterizers are capable of distinguishing between different dynamical regimes and can pick up subtle differences in dynamical behavior, whereas the usual characterizers provide a coarser characterization. However, the two taken in conjunction, can provide information about the dynamical behavior of the time series, as well as the correlations in the evolving system. Our methods can therefore provide powerful tools for the analysis of dynamical systems.

physics.soc-ph↗

arXiv2018

Efficient population transfer via non-ergodic extended states in quantum spin glass

We analyze a new computational role of coherent multi-qubit quantum tunneling that gives rise to bands of non-ergodic extended (NEE) quantum states each formed by a superposition of a large number of computational states (deep local minima of the energy landscape) with similar energies. NEE provide a mechanism for population transfer (PT) between computational states and therefore can serve as a new quantum subroutine for quantum search, quantum parallel tempering and reverse annealing optimization algorithms. We study PT in a quantum n-spin system subject to a transverse field where the energy function $E(z)$ encodes a classical optimization problem over the set of spin configurations $z$. Given an initial spin configuration with low energy, PT protocol searches for other bitstrings at energies within a narrow window around the initial one. We provide an analytical solution for PT in a simple yet nontrivial model: $M$ randomly chosen marked bit-strings are assigned energies $E(z)$ within a narrow strip $[-n -W/2, n + W/2]$, while the rest of the states are assigned energy 0. We find that the scaling of a typical PT runtime with n and L is the same as that in the multi-target Grover's quantum search algorithm, except for a factor that is equal to $\exp(n /(2B^2))$ for finite transverse field $B\gg1$. Unlike the Hamiltonians used in analog quantum unstructured search algorithms known so far, the model we consider is non-integrable and population transfer is not exponentially sensitive in n to the weight of the driver Hamiltonian. We study numerically the PT subroutine as a part of quantum parallel tempering algorithm for a number of examples of binary optimization problems on fully connected graphs.

quant-ph↗

arXiv2018

Frustration induced quasi-many-body localization without disorder

Motivated by the question of whether disorder is a prerequisite for localization to occur in quantum many-body systems, we study a frustrated one-dimensional spin chain, which supports localized many-body eigenstates in the absence of disorder. When the system is prepared in an initial state with one domain wall, it exhibits characteristic signatures of quasi-many-body localization (quasi- MBL), including initial state memory retention, an exponentially increasing lifetime with enlarging the size of the system, a logarithmic growth of entanglement entropy, and a logarithmic light cone of an out-of-time-ordered correlator. We further show that the localized many-body eigenstates can be manipulated as pseudospin-1/2s and thus could potentially serve as qubits. Our findings suggest a new route of using frustration to access quasi-MBL and preserve quantum coherence.

cond-mat.quant-gas↗

arXiv2021

The Physics of Machine Learning: An Intuitive Introduction for the Physical Scientist

This article is intended for physical scientists who wish to gain deeper insights into machine learning algorithms which we present via the domain they know best, physics. We begin with a review of two energy-based machine learning algorithms, Hopfield networks and Boltzmann machines, and their connection to the Ising model. This serves as a foundation to understand the phenomenon of learning more generally. Equipped with this intuition we then delve into additional, more "practical," machine learning architectures including feedforward neural networks, convolutional neural networks, and autoencoders. We also provide code that explicitly demonstrates training a neural network with gradient descent.

cond-mat.dis-nn↗

arXiv2026

A Physics-Regularized Neural Network and Kirchhoff Markov Random Field Framework for Inferring Internal Electrochemical States from Operando Spectromicroscopy

Quantitative understanding of coupled reaction-transport dynamics in lithium-ion battery (LIB) composite electrodes is limited by the inaccessibility of key internal electrochemical states. Here we present a physics-integrated, data-driven framework to reconstruct latent states from operando microscopic X-ray absorption fine structure ($μ$ -XAFS) hyperspectral data of LIB cathodes. Normalized-local-lithium-stoichiometry (NLLS) maps derived from Co K-edge spectra are refined using a physics-regularized neural network that enforces spatial continuity and current conservation to resolve ambiguities in the weak-sensitivity region. The reconstructed NLLS dynamics are embedded in a Kirchhoff-based Markov random field incorporating Kirchhoff's laws, Ohm's law, and Butler-Volmer kinetics to infer interfacial current density, ionic current, electrolyte potential, and effective ionic conductivity. Application to electrodes with differential initial electrolyte concentrations reveals distinct reaction-propagation modes that are consistent with a mechanism involving the non-monotonic concentration dependence of electrolyte conductivity. The inferred electrolyte concentration profiles qualitatively resemble the spatially extended concentration changes observed by independent operando X-ray transmission imaging using a 1 M LiAsF$_6$ electrolyte.

physics.chem-ph↗

arXiv2026

Unifying Physical Backpropagation

Physical computing systems exploit device dynamics for computation, but their gradient-based optimization is challenging: backpropagation through a digital twin suffers from a model-reality gap. On-device gradient computation could resolve this issue, and a handful of theoretical and experimental studies have proposed ways to achieve it. Yet a unifying theory identifying when a physical system can compute the gradient of its own performance has been missing. Here we develop such a unification based on the adjoint method: we identify sufficient conditions under which the adjoint field required for formally exact gradients can be generated on the same hardware that performs the computation. Linear and nonlinear systems obey fundamentally different conditions: for linear systems, damping or gain is admissible provided reciprocity is preserved. For nonlinear trajectory systems, the sufficient conditions are reciprocity of the linearized system and the existence of a time-reversal mirror. Algorithmically, the nonlinear case requires infinitesimal nudging, whereas linear systems admit a finite-amplitude experiment. We recover (quantum) Equilibrium Propagation, Hamiltonian echo backpropagation, fully forward mode training and in situ gradient methods in integrated-photonic and free-space-optical systems. Finally, we show that reciprocity is a special case of more general intertwining conditions. For linear systems, these permit exact on-device gradients in a class of non-Hermitian, non-reciprocal systems. For nonlinear trajectories, they combine with generalized time-reversal mirrors to cover, e.g., PT-symmetric equations. The framework also includes time-dependent parameters and Onsager-reciprocal dynamics, providing a unified basis for formally exact physical learning.

cond-mat.dis-nn↗

arXiv1999

Weak localization in the 2D metallic regime of Si-MOS

The negative magnetoresistance due to weak localization is investigated in the two-dimensional metallic state of Si-MOS structures for high conductance values between 35 and 120 e^2/h. The extracted phase coherence time is equal to the momentum relaxation time at 10 K but nearly 100 times longer at the lowest temperature. Nevertheless, only weak logarithmic corrections to the conductivity are present in the investigated temperature and concentration range thus proving the absence of strong quantum effects due to electron-electron interaction. From saturation effects of the phase coherence time a lower boundary for spin-orbit scattering of about 200 ps is estimated.

cond-mat.mes-hall↗

arXiv2005

Comparative numerical study of Anderson localization in disordered electron systems

Taking into account that a proper description of disordered systems should focus on distribution functions, the authors develop a powerful numerical scheme for the determination of the probability distribution of the local density of states (LDOS), which is based on a Chebyshev expansion with kernel polynomial refinement and allows the study of large finite clusters (up to $100^3$). For the three-dimensional Anderson model it is demonstrated that the distribution of the LDOS shows a significant change at the disorder induced delocalisation-localisation transition. Consequently, the so-called typical density of states, defined as the geometric mean of the LDOS, emerges as a natural order parameter. The calculation of the phase diagram of the Anderson model proves the efficiency and reliability of the proposed approach in comparison to other localisation criteria, which rely, e.g., on the decay of the wavefunction or the inverse participation number.

cond-mat.dis-nn↗

arXiv2013

Optimal scales in weighted networks

The analysis of networks characterized by links with heterogeneous intensity or weight suffers from two long-standing problems of arbitrariness. On one hand, the definitions of topological properties introduced for binary graphs can be generalized in non-unique ways to weighted networks. On the other hand, even when a definition is given, there is no natural choice of the (optimal) scale of link intensities (e.g. the money unit in economic networks). Here we show that these two seemingly independent problems can be regarded as intimately related, and propose a common solution to both. Using a formalism that we recently proposed in order to map a weighted network to an ensemble of binary graphs, we introduce an information-theoretic approach leading to the least biased generalization of binary properties to weighted networks, and at the same time fixing the optimal scale of link intensities. We illustrate our method on various social and economic networks.

physics.data-an↗

arXiv2012

Orbital polarization and magnetization for independent particles in disordered media

Formulas for the contribution of the conduction electrons to the polarization and magnetization are derived for disordered systems and within a one-particle framework. These results generalize known formulas for Bloch electrons and the presented proofs considerably simplify and strengthen prior justifications. The new formulas show that orbital polarization and magnetization are of geometric nature. This leads to quantization for a periodically driven Piezo effect as well as the derivative of the magnetization w.r.t. the chemical potential. It is also shown how the latter is connected to boundary currents in Chern insulators. The main technical tools in the proofs are an adaption of Nenciu's super-adiabatic theory to C$^*$-dynamical systems and Bellissard's Ito derivatives w.r.t. the magnetic field.

math-ph↗

arXiv2012

Complex Random Energy Model: Zeros and Fluctuations

The partition function of the random energy model at inverse temperature $β$ is a sum of random exponentials $Z_N(β)=\sum_{k=1}^N \exp(β\sqrt{n} X_k)$, where $X_1,X_2,...$ are independent real standard normal random variables (= random energies), and $n=\log N$. We study the large $N$ limit of the partition function viewed as an analytic function of the complex variable $β$. We identify the asymptotic structure of complex zeros of the partition function confirming and extending predictions made in the theoretical physics literature. We prove limit theorems for the random partition function at complex $β$, both on the logarithmic scale and on the level of limiting distributions. Our results cover also the case of the sums of independent identically distributed random exponentials with any given correlations between the real and imaginary parts of the random exponent.

math.PR↗

arXiv2013

Low-temperature specific heat of graphite and CeSb2: Validation of a quasi-adiabatic continuous method

We present the application of a fast quasi-adiabatic continuous method to the measurement of specific heat at 4He temperatures, which can be used for the study of a wide range of materials. The technique can be performed in the same configuration used for the relaxation method, as the typical time constants between calorimetric cell and thermal sink at 4.2 K are chosen to be of the order of tau~30 s. The accuracy in the absolute values have been tested by comparing them to relaxation-method results obtained in the same samples (performed in situ using the same set-up), with a deviation between the absolute values < 3% in the whole temperature range. This new version of the continuous calorimetric method at low temperatures allows us to completely characterize and measure a sample within a few hours with a high density of data points, whereas when employing other methods we typically need a few days. An exhaustive study has been performed for reproducibility to be tested. In the present work, we have applied this method to two different substances: CeSb2, which exhibits three magnetic transitions at 15.5 K, 11.7 K and 9.5 K, and graphite, both highly-oriented pyrolytic graphite (HOPG) and natural crystals. Our results on these graphites are discussed in comparison with previous published data on different kinds of graphite samples.

cond-mat.mtrl-sci↗

arXiv2012

Region graph partition function expansion and approximate free energy landscapes: Theory and some numerical results

Graphical models for finite-dimensional spin glasses and real-world combinatorial optimization and satisfaction problems usually have an abundant number of short loops. The cluster variation method and its extension, the region graph method, are theoretical approaches for treating the complicated short-loop-induced local correlations. For graphical models represented by non-redundant or redundant region graphs, approximate free energy landscapes are constructed in this paper through the mathematical framework of region graph partition function expansion. Several free energy functionals are obtained, each of which use a set of probability distribution functions or functionals as order parameters. These probability distribution function/functionals are required to satisfy the region graph belief-propagation equation or the region graph survey-propagation equation to ensure vanishing correction contributions of region subgraphs with dangling edges. As a simple application of the general theory, we perform region graph belief-propagation simulations on the square-lattice ferromagnetic Ising model and the Edwards-Anderson model. Considerable improvements over the conventional Bethe-Peierls approximation are achieved. Collective domains of different sizes in the disordered and frustrated square lattice are identified by the message-passing procedure. Such collective domains and the frustrations among them are responsible for the low-temperature glass-like dynamical behaviors of the system.

cond-mat.dis-nn↗