Searcharxiv⌕ Search

arXiv subjects

Xiaoyu Xie

Publications and source records attributed to Xiaoyu Xie.

At least 19 recordsLinked to original sources
arXiv2026

StoryEcho: A Narrative Mirroring Loop Generative Storytelling System for Picky-Eating Intervention

Picky eating can limit children's dietary variety and create tension in family feeding routines. Existing food-related technologies often focus on mealtime intervention or standalone educational artifacts, offering limited support for connecting low-pressure narrative engagement with children's real-world food exploration over time. We present StoryEcho, a generative storytelling system centered on a narrative mirroring loop, in which personalized stories model sensory exploration through a persistent counterpart and children's subsequent food encounters are reflected back into narrative feedback and future story development. Informed by a formative study, we designed StoryEcho and evaluated it in a 14-day between-subjects field study with 26 families. Compared with food-personalized generative stories without the narrative mirroring loop, StoryEcho was associated with higher try level, approach, and intake, and lower resistance and caregiver pressure. These findings suggest that narrative mirroring can support children's low-pressure food exploration in family routines, while highlighting design tensions for future generative storytelling interventions.

cs.HC↗
arXiv2026

Reconstruction and Reflection of Positive Experiences through Resurfacing Laughter-indexed Everyday Moments

Positive everyday moments often escape deliberate recording, while continuous self-tracking can generate extensive records that are difficult to revisit. We explore laughter as a naturally occurring, sparse index for constructing contextualized personal records to support later reconstruction and reflection. A formative study with 12 participants characterized laughter as an affective but semantically incomplete index and informed \textit{LaughAnchor}, a mobile and wearable self-tracking system. During participant-initiated recording, the system assembles detected laughter and aligned context into candidate moments for later reconstruction and reflection, with layered context disclosure, user-controlled curation, and near-term and long-term resurfacing. In a three-week field deployment with 12 participants, passive indexing preserved moments they considered unlikely to record deliberately but valued retrospectively. During resurfacing, participants attributed affective re-experiencing to laughter and used additional context both to reconstruct episodes and to explore already-recalled experiences. Across moments and reviews, resurfacing supported rediscovery and broader awareness of relationships, routines, and emotional states. These findings inform self-tracking designs that use sparse affective indices to organize contextual records for reconstruction and reflection, while keeping interpretation and retention under user control.

cs.HC↗
arXiv2025

A Tutorial on Dimensionless Learning: Geometric Interpretation and the Effect of Noise

Dimensionless learning is a data-driven framework for discovering dimensionless numbers and scaling laws from experimental measurements. This tutorial introduces the method, explaining how it transforms experimental data into compact physical laws that reveal compact dimensional invariance between variables. The approach combines classical dimensional analysis with modern machine learning techniques. Starting from measurements of physical quantities, the method identifies the fundamental ways to combine variables into dimensionless groups, then uses neural networks to discover which combinations best predict the experimental output. A key innovation is a regularization technique that encourages the learned coefficients to take simple, interpretable values like integers or half-integers, making the discovered laws both accurate and physically meaningful. We systematically investigate how measurement noise and discrete sampling affect the discovery process, demonstrating that the regularization approach provides robustness to experimental uncertainties. The method successfully handles cases with single or multiple dimensionless numbers, revealing how different but equivalent representations can capture the same underlying physics. Despite recent progress, key challenges remain, including managing the computational cost of identifying multiple dimensionless groups, understanding the influence of data characteristics, automating the selection of relevant input variables, and developing user-friendly tools for experimentalists. This tutorial serves as both an educational resource and a practical guide for researchers seeking to apply dimensionless learning to their experimental data.

cs.LG↗
arXiv2025

Convolutional Hierarchical Deep Learning Neural Networks-Tensor Decomposition (C-HiDeNN-TD): a scalable surrogate modeling approach for large-scale physical systems

A common trend in simulation-driven engineering applications is the ever-increasing size and complexity of the problem, where classical numerical methods typically suffer from significant computational time and huge memory cost. Methods based on artificial intelligence have been extensively investigated to accelerate partial differential equations (PDE) solvers using data-driven surrogates. However, most data-driven surrogates require an extremely large amount of training data. In this paper, we propose the Convolutional Hierarchical Deep Learning Neural Network-Tensor Decomposition (C-HiDeNN-TD) method, which can directly obtain surrogate models by solving large-scale space-time PDE without generating any offline training data. We compare the performance of the proposed method against classical numerical methods for extremely large-scale systems.

cs.CE↗
arXiv2025

Fluctuations in Quantum Unique Ergodicity at the Spectral Edge

We study the eigenvector mass distribution of an $N\times N$ Wigner matrix on a set of coordinates $I$ satisfying $| I | \ge c N$ for some constant $c >0$. For eigenvectors corresponding to eigenvalues at the spectral edge, we show that the sum of the mass on these coordinates converges to a Gaussian in the $N \rightarrow \infty$ limit, after a suitable rescaling and centering. The proof proceeds by a two moment matching argument. We directly compare edge eigenvector observables of an arbitrary Wigner matrix to those of a Gaussian matrix, which may be computed explicitly.

math.PR↗
arXiv2025

Interpolating Neural Network-Tensor Decomposition (INN-TD): a scalable and interpretable approach for large-scale physics-based problems

Deep learning has been extensively employed as a powerful function approximator for modeling physics-based problems described by partial differential equations (PDEs). Despite their popularity, standard deep learning models often demand prohibitively large computational resources and yield limited accuracy when scaling to large-scale, high-dimensional physical problems. Their black-box nature further hinders the application in industrial problems where interpretability and high precision are critical. To overcome these challenges, this paper introduces Interpolating Neural Network-Tensor Decomposition (INN-TD), a scalable and interpretable framework that has the merits of both machine learning and finite element methods for modeling large-scale physical systems. By integrating locally supported interpolation functions from finite element into the network architecture, INN-TD achieves a sparse learning structure with enhanced accuracy, faster training/solving speed, and reduced memory footprint. This makes it particularly effective for tackling large-scale high-dimensional parametric PDEs in training, solving, and inverse optimization tasks in physical problems where high precision is required.

cs.CE↗
arXiv2024

Interpolating neural network: A novel unification of machine learning and interpolation theory

Artificial intelligence (AI) has revolutionized software development, shifting from task-specific codes (Software 1.0) to neural network-based approaches (Software 2.0). However, applying this transition in engineering software presents challenges, including low surrogate model accuracy, the curse of dimensionality in inverse design, and rising complexity in physical simulations. We introduce an interpolating neural network (INN), grounded in interpolation theory and tensor decomposition, to realize Engineering Software 2.0 by advancing data training, partial differential equation solving, and parameter calibration. INN offers orders of magnitude fewer trainable/solvable parameters for comparable model accuracy than traditional multi-layer perceptron (MLP) or physics-informed neural networks (PINN). Demonstrated in metal additive manufacturing, INN rapidly constructs an accurate surrogate model of Laser Powder Bed Fusion (L-PBF) heat transfer simulation, achieving sub-10-micrometer resolution for a 10 mm path in under 15 minutes on a single GPU. This makes a transformative step forward across all domains essential to engineering software.

cs.LG↗
arXiv2024

Efficient simulation of inhomogeneously correlated systems using block interaction product states

The strength of DMRG lies in its treatment of identical sites that are energetically degenerate and spatially similar. However, this becomes a drawback when applied to quantum chemistry calculations for large systems, as entangled orbitals often span broad ranges in energy and space, with notably inhomogeneous interactions. In this study, we propose addressing strong intra-fragment and weak inter-fragment correlations separately using a multi-configurational block interaction product state (BIPS) framework. The strong correlation is captured in electronic states on fragments, considering entanglement between fragments and their environments. This method has been tested in various chemical systems and shows high accuracy and efficiency in addressing inhomogeneous effects in quantum chemistry.

quant-ph↗
arXiv2024

The $\ell_r$-Levy-Grothendieck problem and $r\rightarrow p$ norms of Levy matrices

Given an $n\times n$ matrix $A_n$ and $1\leq r, p \leq\infty$, consider the following quadratic optimization problem referred to as the $\ell_r$-Grothendieck problem: \begin{align}M_r(A_n)\coloneqq\max_{\boldsymbol{x}\in\mathbb{R}^n:\|\boldsymbol{x}\|_r\leq1}\boldsymbol{x}^{\top} A_n \boldsymbol{x},\end{align} as well as the $r\rightarrow p$ operator norm of the matrix $A_n$, defined as \begin{align}\|A_n\|_{r \rightarrow p}\coloneqq \sup _{\boldsymbol{x}\in\mathbb{R}^n:\|\boldsymbol{x}\|_r \leq 1}\|A_n \boldsymbol{x}\|_p,\end{align} where $\|\boldsymbol{x}\|_r$ denotes the $\ell_r$-norm of the vector $\boldsymbol{x}$. This work analyzes high-dimensional asymptotics of these quantities when $A_n$ are symmetric random matrices with independent and identically distributed heavy-tailed upper-triangular entries with index $α$. When $1\leq r\leq 2$ (respectively, $1\leq r\leq p$) and $α\in(0,2)$, suitably scaled versions of $M_r(A_n)$ and $\|A_n\|_{r\rightarrow p}$ are shown to converge to a Fréchet distribution as $n\rightarrow\infty$. In contrast, when $2< r<\infty$ (respectively, $1\leq p< r$), it is shown that there exists $α_*\in(1,2)$ such that for every $α\in(0,α_*)$, suitably scaled versions of $M_r(A_n)$ and $\|A_n\|_{r\rightarrow p}$ converge to the power of a stable distribution. Furthermore, it is shown that there exists $\barα_*>α_*$ such that when $α\in(α_*,\barα_*)$, the latter convergence result holds only when the matrix entries are centered; when the entries have non-zero mean, a different limit arises after additional centering and scaling. As a corollary, these results yield a characterization of the limiting ground state of the Levy spin glass when $α\in (0,1)$. The analysis uses a combination of tools from the theory of heavy-tailed distributions, the nonlinear power method and concentration inequalities.

math.PR↗
arXiv2024

Smooth and Sparse Latent Dynamics in Operator Learning with Jerk Regularization

Spatiotemporal modeling is critical for understanding complex systems across various scientific and engineering disciplines, but governing equations are often not fully known or computationally intractable due to inherent system complexity. Data-driven reduced-order models (ROMs) offer a promising approach for fast and accurate spatiotemporal forecasting by computing solutions in a compressed latent space. However, these models often neglect temporal correlations between consecutive snapshots when constructing the latent space, leading to suboptimal compression, jagged latent trajectories, and limited extrapolation ability over time. To address these issues, this paper introduces a continuous operator learning framework that incorporates jerk regularization into the learning of the compressed latent space. This jerk regularization promotes smoothness and sparsity of latent space dynamics, which not only yields enhanced accuracy and convergence speed but also helps identify intrinsic latent space coordinates. Consisting of an implicit neural representation (INR)-based autoencoder and a neural ODE latent dynamics model, the framework allows for inference at any desired spatial or temporal resolution. The effectiveness of this framework is demonstrated through a two-dimensional unsteady flow problem governed by the Navier-Stokes equations, highlighting its potential to expedite high-fidelity simulations in various scientific and engineering applications.

cs.LG↗
arXiv2023

Statistical Parameterized Physics-Based Machine Learning Digital Twin Models for Laser Powder Bed Fusion Process

A digital twin (DT) is a virtual representation of physical process, products and/or systems that requires a high-fidelity computational model for continuous update through the integration of sensor data and user input. In the context of laser powder bed fusion (LPBF) additive manufacturing, a digital twin of the manufacturing process can offer predictions for the produced parts, diagnostics for manufacturing defects, as well as control capabilities. This paper introduces a parameterized physics-based digital twin (PPB-DT) for the statistical predictions of LPBF metal additive manufacturing process. We accomplish this by creating a high-fidelity computational model that accurately represents the melt pool phenomena and subsequently calibrating and validating it through controlled experiments. In PPB-DT, a mechanistic reduced-order method-driven stochastic calibration process is introduced, which enables the statistical predictions of the melt pool geometries and the identification of defects such as lack-of-fusion porosity and surface roughness, specifically for diagnostic applications. Leveraging data derived from this physics-based model and experiments, we have trained a machine learning-based digital twin (PPB-ML-DT) model for predicting, monitoring, and controlling melt pool geometries. These proposed digital twin models can be employed for predictions, control, optimization, and quality assurance within the LPBF process, ultimately expediting product development and certification in LPBF-based metal additive manufacturing.

cs.LG↗
arXiv2023

Quenched large deviation principles for random projections of $\ell_p^n$ balls

Let $(k_n)_{n \in \mathbb{N}}$ be a sequence of positive integers growing to infinity at a sublinear rate, $k_n \rightarrow \infty$ and $k_n/n \rightarrow 0$ as $n \rightarrow \infty$. Given a sequence of $n$-dimensional random vectors $\{Y^{(n)}\}_{n \in \mathbb{N}}$ belonging to a certain class, which includes uniform distributions on suitably scaled $\ell_p^n$-balls or $\ell_p^n$-spheres, $p \geq 2$, and product distributions with sub-Gaussian marginals, we study the large deviations behavior of the corresponding sequence of $k_n$-dimensional orthogonal projections $n^{-1/2} \boldsymbol{a}_{n,k_n} Y^{(n)}$, where $\boldsymbol{a}_{n,k_n}$ is an $(n \times k_n)$-dimensional projection matrix lying in the Stiefel manifold of orthonormal $k_n$-frames in $\mathbb{R}^n$. For almost every sequence of projection matrices, we establish a large deviation principle (LDP) for the corresponding sequence of projections, with a fairly explicit rate function that does not depend on the sequence of projection matrices. As corollaries, we also obtain quenched LDPs for sequences of $\ell_2$-norms and $\ell_\infty$-norms of the coordinates of the projections. Past work on LDPs for projections with growing dimension has mainly focused on the annealed setting, where one also averages over the random projection matrix, chosen from the Haar measure, in which case the coordinates of the projection are exchangeable. The quenched setting lacks such symmetry properties, and gives rise to significant new challenges in the setting of growing projection dimension. Along the way, we establish new Gaussian approximation results on the Stiefel manifold that may be of independent interest. Such LDPs are of relevance in asymptotic convex geometry, statistical physics and high-dimensional statistics.

math.PR↗
arXiv2022

JAX-FEM: A differentiable GPU-accelerated 3D finite element solver for automatic inverse design and mechanistic data science

This paper introduces JAX-FEM, an open-source differentiable finite element method (FEM) library. Constructed on top of Google JAX, a rising machine learning library focusing on high-performance numerical computing, JAX-FEM is implemented with pure Python while scalable to efficiently solve problems with moderate to large sizes. For example, in a 3D tensile loading problem with 7.7 million degrees of freedom, JAX-FEM with GPU achieves around 10$\times$ acceleration compared to a commercial FEM code depending on platform. Beyond efficiently solving forward problems, JAX-FEM employs the automatic differentiation technique so that inverse problems are solved in a fully automatic manner without the need to manually derive sensitivities. Examples of 3D topology optimization of nonlinear materials are shown to achieve optimal compliance. Finally, JAX-FEM is an integrated platform for machine learning-aided computational mechanics. We show an example of data-driven multi-scale computations of a composite material where JAX-FEM provides an all-in-one solution from microscopic data generation and model training to macroscopic FE computations. The source code of the library and these examples are shared with the community to facilitate computational mechanics research.

cs.MS↗
arXiv2021

Stochastic Adaptive Single-Site Time-Dependent Variational Principle

In recent years, the time-dependent variational principle (TDVP) method based on the matrix product state (MPS) wave function formulation has shown its great power in performing large-scale quantum dynamics simulations for realistic chemical systems with strong electron-vibration interactions. In this work, we propose a new stochastic adaptive single-site TDVP (SA-1TDVP) scheme to evolve the bond-dimension adaptively, which can integrate the tra-ditional advantages of both the high efficiency of single-site TDVP (1TDVP) variant and the high accuracy of the two-site TDVP (2TDVP) variant. Based on the assumption that the level statistics of entanglement Hamiltonians, which originate from the reduced density matrices of the MPS method, follows a Poisson or Wigner distribution, as generically predicted by random matrix theory, addi-tional random singular values are generated to expand the bond-dimension automatically. Tests on simulating the vibrationally-resolved quantum dynamics and absorption spectra in the pyrazine molecule and perylene bisimide (PBI) J-aggregate trimer as well as a spin-1/2 Heisenberg chain show that it can be automatic and as accurate as 2TDVP but reduce the computational time remarkably.

cond-mat.str-el↗
arXiv2021

Data-driven discovery of dimensionless numbers and scaling laws from experimental measurements

Dimensionless numbers and scaling laws provide elegant insights into the characteristic properties of physical systems. Classical dimensional analysis and similitude theory fail to identify a set of unique dimensionless numbers for a highly-multivariable system with incomplete governing equations. In this study, we embed the principle of dimensional invariance into a two-level machine learning scheme to automatically discover dominant and unique dimensionless numbers and scaling laws from data. The proposed methodology, called dimensionless learning, can reduce high-dimensional parametric spaces into descriptions involving just a few physically-interpretable dimensionless parameters, which significantly simplifies the process design and optimization of the system. We demonstrate the algorithm by solving several challenging engineering problems with noisy experimental measurements (not synthetic data) collected from the literature. The examples include turbulent Rayleigh-Benard convection, vapor depression dynamics in laser melting of metals, and porosity formation in 3D printing. We also show that the proposed approach can identify dimensionally-homogeneous differential equations with minimal parameters by leveraging sparsity-promoting techniques.

physics.flu-dyn↗
arXiv2020

Charge transfer via deep hole in the J51/N2200 blend

In recently developed non-fullerene acceptor (NFA) based organic solar cells (OSCs), both the donor and acceptor parts can be excited by absorbing light photons. Therefore, both electron transfer and hole transfer channels could occur at the donor/acceptor interface for generating free charge carriers in NFA based OSCs. However, in many molecular and DNA systems, recent studies revealed the high charge transfer (CT) efficiency cannot be reasonably explained by a CT model with only highest occupied molecular orbitals (HOMOs) and lowest unoccupied molecular orbitals (LUMOs) of donor and acceptor molecules. In this work, taking an example of a full-polymer blend consisting of benzodithiophenealt-benzotriazole copolymers (J51) as donor and naphthalene diimide-bithiophene (N2200) as acceptor, in which the ultrafast hole transfer has been recently reported, we investigate its CT process and examine the different roles of various frontier molecular orbitals. Through a joint study of quantum mechanics electronic structure calculation and nonadiabatic dynamics simulation, we find the hole transfer between HOMOs of J51 and N2200 can hardly happen but the hole transfer from HOMO of N2200 to HOMO-1 of J51 is much more efficient. This points out the underlying importance of deep hole channel in CT process and indicates that including frontier molecular orbitals (FMOs) other than HOMOs and LUMOs is highly necessary to build a robust physical model for studying CT process in molecular optoelectronic materials.

physics.chem-ph↗
arXiv2019

Time-dependent Density Matrix Renormalization Group Quantum Dynamics for Realistic Chemical Systems

Electronic and/or vibronic coherence has been found by recent ultrafast spectroscopy experiments in many chemical, biological and material systems. This indicates that there are strong and complicated interactions between electronic states and vibration modes in realistic chemical systems. Therefore, simulations of quantum dynamics with a large number of electronic and vibrational degrees of freedom are highly desirable. Due to the efficient compression and localized representation of quantum states in the matrix-product state (MPS) formulation, time-evolution methods based on the MPS framework, which we summarily refer to as tDMRG (time-dependent density-matrix renormalization group) methods, are considered to be promising candidates to study the quantum dynamics of realistic chemical systems. In this work, we benchmark the performances of four different tDMRG methods, including global Taylor, global Krylov, local one-site and two-site time-dependent variational principle (1TDVP and 2TDVP), with a comparison to multi-configuration time-dependent Hartree (MCTDH) and experimental results. Two typical chemical systems of internal conversion and singlet fission are investigated, one containing strong and high-order local and non-local electron-vibration couplings, the other exhibiting a continuous phonon bath. The comparison shows that the tDMRG methods (particularly, the 2TDVP method) can describe the full quantum dynamics in large chemical systems accurately and efficiently. Several key parameters in the tDMRG calculation including the truncation error threshold, time interval and ordering of local sites were also investigated to strike the balance between efficiency and accuracy of results.

cond-mat.str-el↗