Search

High-entropy materials shift the traditional materials discovery paradigm to one that leverages disorder, enabling access to unique chemistries unreachable through enthalpy alone. We present a self-consistent approach integrating computation and experiment to understand and explore single-phase rock salt high-entropy oxides. By leveraging a machine-learning interatomic potential, we rapidly and accurately map high-entropy composition space using our two descriptors: bond length distribution and mixing enthalpy. The single-phase stabilities for all experimentally stabilized rock salt compositions are correctly resolved, with dozens more compositions awaiting discovery.

High entropy oxides (HEO)s have garnered much interest due to their available high degree of tunability. Here, we study the local structure of (MgNiCuCoZn)0.167(MnCr)0.083O, a composition based on the parent HEO (MgNiCuCoZn)0.2O.We synthesized a series of thin films via pulsed laser deposition at incremental oxygen partial pressures. X-ray diffraction shows lattice parameter to decrease with increased pO2 pressures until the onset of phase separation. X-ray absorption fine structure shows that specific atomic species in the composition dictate the global structure of the material as Cr, Co, and Mn shift to energetically favorable coordination with increasing pressure. Transmission electron microscopy analysis on a lower-pressure sample exhibits a rock salt structure, but the higher-pressure sample reveals reflections reminiscent of the spinel structure. In all, these findings give a more complete picture on how (MgNiCuCoZn)0.167(MnCr)0.083O forms with varying initial conditions and advances fundamental knowledge of cation behavior in high entropy oxides.

We unravel the distinct roles each cation plays in phase evolution, stability, and properties within Mg1/5Co1/5Ni1/5Cu1/5Zn1/5O high-entropy oxide (HEO) by integrating experimental findings, thermodynamic analyses, and first-principles predictions. Our approach is through sequentially removing one cation at a time from the five-component high-entropy oxide to create five four-component derivatives. Bulk synthesis experiments indicate that Mg, Ni, and Co act as rock salt phase stabilizers whereas only Mg and Ni enthalpically enhance single-phase rock salt stability in thin film growth; synthesis conditions dictate whether Co is a rock salt phase stabilizer or destabilizer. By examining the competing phases and oxidation state preferences using pseudo-binary phase diagrams and first-principles calculations, we resolve the stability differences between bulk and thin film for all compositions. We systematically explore HEO macroscopic property sensitivity to cation selection employing both predicted and measured optical spectra. This study establishes a framework for understanding high-entropy oxide synthesizability and properties on a per-cation basis that is broadly applicable to tailoring functional property design in other high-entropy materials., Comment: Saeed S. I. Almishal and Jacob T. Sivak contributed equally to this work

We report on the structure and dielectric properties of ternary A6B2O17 (A = Zr; B = Nb, Ta) thin films and ceramics. Thin films are produced via sputter deposition from dense, phase-homogenous bulk ceramic targets, which are synthesized through a reactive sintering process at 1500 {\deg}C. Crystal structure, microstructure, chemistry and dielectric properties are characterized by X-ray diffraction and reflectivity, atomic force microscopy, X-ray photoelectron spectroscopy, and capacitance analysis respectively. We observe relative permittivities approaching 60 and loss tangents < 10^-2 across the 10^3-10^5 Hz frequency range in the Zr6Nb2O17 and Zr6Ta2O17 phases. These observations create an opportunity space for this novel class of disordered oxide electroceramics., Comment: Journal of the American Ceramic Society, 2024

Interest in high-entropy inorganic compounds originates from their ability to stabilize cations and anions in local environments that rarely occur at standard temperature and pressure. This leads to new crystalline phases in many-cation formulations with structures and properties that depart from conventional trends. The highest-entropy homogeneous and random solid-solution is a parent structure from which a continuum of lower-entropy offspring can originate by adopting chemical and/or structural order. This report demonstrates how synthesis conditions, thermal history, and elastic and chemical boundary conditions conspire to regulate this process in Mg0.2Co0.2Ni0.2Cu0.2Zn0.2O, during which coherent CuO nano-tweeds and spinel nano-cuboids evolve. We do so by combining structured synthesis routes, atomic-resolution microscopy and spectroscopy, density functional theory, and a phase field modeling framework that accurately predicts the emergent structure and local chemistry. This establishes a framework to appreciate, understand, and predict the macrostate spectrum available to a high-entropy system that is critical to rationalize property engineering opportunities.

Convolutional neural networks (CNNs) are one of the most widely used neural network architectures, showcasing state-of-the-art performance in computer vision tasks. Although larger CNNs generally exhibit higher accuracy, their size can be effectively reduced by ``tensorization'' while maintaining accuracy, namely, replacing the convolution kernels with compact decompositions such as Tucker, Canonical Polyadic decompositions, or quantum-inspired decompositions such as matrix product states, and directly training the factors in the decompositions to bias the learning towards low-rank decompositions. But why doesn't tensorization seem to impact the accuracy adversely? We explore this by assessing how \textit{truncating} the convolution kernels of \textit{dense} (untensorized) CNNs impact their accuracy. Specifically, we truncated the kernels of (i) a vanilla four-layer CNN and (ii) ResNet-50 pre-trained for image classification on CIFAR-10 and CIFAR-100 datasets. We found that kernels (especially those inside deeper layers) could often be truncated along several cuts resulting in significant loss in kernel norm but not in classification accuracy. This suggests that such ``correlation compression'' (underlying tensorization) is an intrinsic feature of how information is encoded in dense CNNs. We also found that aggressively truncated models could often recover the pre-truncation accuracy after only a few epochs of re-training, suggesting that compressing the internal correlations of convolution layers does not often transport the model to a worse minimum. Our results can be applied to tensorize and compress CNN models more effectively., Comment: 40 pages, 21 images

Large Language Models (LLMs) such as ChatGPT and LlaMA are advancing rapidly in generative Artificial Intelligence (AI), but their immense size poses significant challenges, such as huge training and inference costs, substantial energy demands, and limitations for on-site deployment. Traditional compression methods such as pruning, distillation, and low-rank approximation focus on reducing the effective number of neurons in the network, while quantization focuses on reducing the numerical precision of individual weights to reduce the model size while keeping the number of neurons fixed. While these compression methods have been relatively successful in practice, there is no compelling reason to believe that truncating the number of neurons is an optimal strategy. In this context, this paper introduces CompactifAI, an innovative LLM compression approach using quantum-inspired Tensor Networks that focuses on the model's correlation space instead, allowing for a more controlled, refined and interpretable model compression. Our method is versatile and can be implemented with - or on top of - other compression techniques. As a benchmark, we demonstrate that a combination of CompactifAI with quantization allows to reduce a 93% the memory size of LlaMA 7B, reducing also 70% the number of parameters, accelerating 50% the training and 25% the inference times of the model, and just with a small accuracy drop of 2% - 3%, going much beyond of what is achievable today by other compression techniques. Our methods also allow to perform a refined layer sensitivity profiling, showing that deeper layers tend to be more suitable for tensor network compression, which is compatible with recent observations on the ineffectiveness of those layers for LLM performance. Our results imply that standard LLMs are, in fact, heavily overparametrized, and do not need to be large at all., Comment: 5 pages, 4 figures, 2 tables, and supplementary information of 2 pages and 1 figure. Revised version with new benchmarks for LlaMA2-7B

We show how quantum-inspired 2d tensor networks can be used to efficiently and accurately simulate the largest quantum processors from IBM, namely Eagle (127 qubits), Osprey (433 qubits) and Condor (1121 qubits). We simulate the dynamics of a complex quantum many-body system -- specifically, the kicked Ising experiment considered recently by IBM in Nature 618, p. 500-505 (2023) -- using graph-based Projected Entangled Pair States (gPEPS), which was proposed by some of us in PRB 99, 195105 (2019). Our results show that simple tensor updates are already sufficient to achieve very large unprecedented accuracy with remarkably low computational resources for this model. Apart from simulating the original experiment for 127 qubits, we also extend our results to 433 and 1121 qubits, and for evolution times around 8 times longer, thus setting a benchmark for the newest IBM quantum machines. We also report accurate simulations for infinitely-many qubits. Our results show that gPEPS are a natural tool to efficiently simulate quantum computers with an underlying lattice-based qubit connectivity, such as all quantum processors based on superconducting qubits., Comment: 7 pages, 8 figures, revised version

We map out the ground state phase diagram of the isotropic Kekul{\'e}-Kitaev model on the honeycomb lattice in the presence of the Heisenberg exchange couplings. Our study relies on large-scale tensor network simulations based on graph-based projected entangled pair state (gPEPS) approach in the thermodynamic limit. We find that on top of the quantum spin liquid (QSL) and conventional magnetically ordered phases which are typical of the Kitaev-Heisenberg model, the Kekul{\'e}-Heisenberg phase diagram, hosts two plaquette valance bond solid (VBS) phases with vanishing magnetic order. While the VBS phases preserve the symmetries of the original Hamiltonian, they differ markedly from the Kitaev spin liquid by having decorated plaquette ordering which is distinguished by a plaquette order parameter.

Recent advances in deep learning have enabled us to address the curse of dimensionality (COD) by solving problems in higher dimensions. A subset of such approaches of addressing the COD has led us to solving high-dimensional PDEs. This has resulted in opening doors to solving a variety of real-world problems ranging from mathematical finance to stochastic control for industrial applications. Although feasible, these deep learning methods are still constrained by training time and memory. Tackling these shortcomings, Tensor Neural Networks (TNN) demonstrate that they can provide significant parameter savings while attaining the same accuracy as compared to the classical Dense Neural Network (DNN). In addition, we also show how TNN can be trained faster than DNN for the same accuracy. Besides TNN, we also introduce Tensor Network Initializer (TNN Init), a weight initialization scheme that leads to faster convergence with smaller variance for an equivalent parameter count as compared to a DNN. We benchmark TNN and TNN Init by applying them to solve the parabolic PDE associated with the Heston model, which is widely used in financial pricing theory., Comment: 11 pages, 8 figures, minor changes. arXiv admin note: substantial text overlap with arXiv:2208.02235

Deep neural networks (NNs) encounter scalability limitations when confronted with a vast array of neurons, thereby constraining their achievable network depth. To address this challenge, we propose an integration of tensor networks (TN) into NN frameworks, combined with a variational DMRG-inspired training technique. This in turn, results in a scalable tensor neural network (TNN) architecture capable of efficient training over a large parameter space. Our variational algorithm utilizes a local gradient-descent technique, enabling manual or automatic computation of tensor gradients, facilitating design of hybrid TNN models with combined dense and tensor layers. Our training algorithm further provides insight on the entanglement structure of the tensorized trainable weights and correlation among the model parameters. We validate the accuracy and efficiency of our method by designing TNN models and providing benchmark results for linear and non-linear regressions, data classification and image recognition on MNIST handwritten digits., Comment: 21 pages, 19 figures

Partial Differential Equations (PDEs) are used to model a variety of dynamical systems in science and engineering. Recent advances in deep learning have enabled us to solve them in a higher dimension by addressing the curse of dimensionality in new ways. However, deep learning methods are constrained by training time and memory. To tackle these shortcomings, we implement Tensor Neural Networks (TNN), a quantum-inspired neural network architecture that leverages Tensor Network ideas to improve upon deep learning approaches. We demonstrate that TNN provide significant parameter savings while attaining the same accuracy as compared to the classical Dense Neural Network (DNN). In addition, we also show how TNN can be trained faster than DNN for the same accuracy. We benchmark TNN by applying them to solve parabolic PDEs, specifically the Black-Scholes-Barenblatt equation, widely used in financial pricing theory, empirically showing the advantages of TNN over DNN. Further examples, such as the Hamilton-Jacobi-Bellman equation, are also discussed., Comment: 14 pages, 11 figures, minimal changes

Wireless communication in millimeter wave spectrum is poised to provide the latency and bandwidth needed for advanced use cases unfeasible at lower frequencies. Despite the market potential of vehicular communication networks, investigations into the millimeter wave vehicular channel are lacking. In this paper, we present a detailed overview of a novel 1 GHz wide, multi-antenna vehicle to vehicle directional channel sounding and measurement platform operating at 28 GHz. The channel sounder uses two 256-element phased arrays at the transmitter vehicle and four 64-element arrays at the receiver vehicle, with the receiver measuring 116 different directional beams in less than 1 millisecond. By measuring the full multi-beam channel impulse response at large bandwidths, our system provides unprecedented insight in instantaneous mobile vehicle to vehicle channels. The system also uses centimeter-level global position tracking and 360 degree video capture to provide additional contextual information for joint communication and sensing applications. An initial measurement campaign was conducted on highway and surface streets in Austin, Texas. We show example data that highlights the sensing capability of the system. Preliminary results from the measurement campaign show that bumper mounted mmWave arrays provide rich scattering in traffic as well a provide significant directional diversity aiding towards high reliability vehicular communication. Additionally, potential waveguide effects from high traffic in lanes can also extend the range of mmWave signals significantly., Comment: 2022 IEEE Wireless Communications and Networking Conference (WCNC)

We demonstrate power-efficient, thermo-optic, silicon nitride waveguide phase shifters for blue, green, and yellow wavelengths. The phase shifters operated with low power consumption due to a suspended structure and multi-pass waveguide design. The devices were fabricated on 200-mm silicon wafers using deep ultraviolet lithography as part of an active visible-light integrated photonics platform. The measured power consumption to achieve a $\pi$ phase shift (averaged over multiple devices) was 0.78, 0.93, 1.09, and 1.20 mW at wavelengths of 445, 488, 532, and 561 nm, respectively. The phase shifters were integrated into Mach-Zehnder interferometer switches, and $10- 90$\% rise(fall) times of about 570(590) $\mu$s were measured., Comment: Zheng Yong and Hong Chen contributed equally to this work

We combine tensor-network approaches and high-order linked-cluster expansions to investigate the quantum phase diagram of the antiferromagnetic Kitaev's honeycomb model in a magnetic field for general spin values. For the pure Kitaev model, tensor network calculations confirm the absence of fluxes and spin-spin correlations beyond nearest neighbor in the ground state, but signal a breaking of the discrete orientational symmetry for $S\in\{1,3/2,2\}$ inline with the semiclassical limit. An intermediate region between Kitaev phases and the high-field polarized phase is demonstrated for all considered spin values. In this intermediate region the tensor network results display a sequence of potential phases whose number increases with the spin value. Each of these can be characterized by distinct local magnetization patterns while the total magnetization increases smoothly as a function of the field. The analysis of the high-field zero-momentum gap and the associated spectral weight of the polarized phase for general spin $S$ obtained by linked-cluster expansions is consistent with an unconventional quantum critical breakdown of the high-field polarized phase in accordance with the presence of exotic physics at intermediate Kitaev couplings., Comment: 23 pages, 15 figures