Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,742 papers · 148 categories

Trend · papers per month

25.0%50.0%75.0%100.0% · Sep 199219922001200920172026
48 results for differentiable physics

Develops scalable differentiable physics for complex object interactions.

problem Limited scalability of existing differentiable physics solvers.
method Adopting meshes for arbitrary geometry, localized collision handling, and accelerated implicit differentiation.
result Significantly reduces memory and computation requirements compared to particle-based methods.

ξ-torch simplifies physics-informed learning by providing differentiable functionals.

problem Training physics-informed deep neural networks requires differentiable physical simulations.
method ξ-torch offers a library of differentiable functionals for scientific simulations.
result Improves numerical stability and reduces memory requirements for higher order derivatives.

We introduce the historical development and physical idea behind topological Yang-Mills theory and explain how a physical framework describing subatomic physics can be used as a tool to study differential geometry. Further, we emphasize that this phenomenon demonstrates that the interrelation between physics and mathem…

2008-03-10abs ↗pdf ↗

This work integrates differentiation and integration in Physics-Informed Neural Networks.

problem Solving integro-differential equations and computing integral transforms.
method Augmenting Physics-Informed Neural Networks with automatic integration.
result Solving complex integral transforms and integro-differential equations.

Unified bounds for neural networks incorporating physical laws.

problem Limitations in existing generalization analyses for PINNs and VPINNs.
method Unified framework using Taylor expansion and Koopman-based analysis.
result High-rank networks can generalize well even with differential operators.

JAX MD enables differentiable physics simulations for molecular dynamics.

problem Performing efficient and differentiable physics simulations for molecular dynamics.
method Differentiable physics simulation environments, interaction potentials, neural networks, flexible primitives.
result Differentiable physics simulations can be used for meta-optimization and scaling to large particle systems.

Gradient estimation techniques applied to programs with randomness in high energy physics.

problem Differentiating programs with discrete randomness in high energy physics.
method Several gradient estimation techniques, including Stochastic AD method, applied to simplified detector design experiments.
result Development of the first fully differentiable branching program.

Bayesian model learns physics laws from data with uncertainty quantification.

problem Lack of uncertainty in discovering governing physical laws from data.
method Bayesian approach with leaf and root modules, Gaussian process for operators, automatic differentiation.
result Quantifies reliability of learned physics laws and propagates uncertainty.

Three types of equations of mathematical physics, namely, the equations, which describe any physical processes, the equations of mechanics and physics of continuous media, and field-theory equations are studied in this paper. In the first and second case the investigation is reduced to the analysis of the nonidentical …

2007-02-06abs ↗pdf ↗

New method solves differential equations on manifolds, with applications in physics.

problem Solving differential equations on Riemannian manifolds.
method Developed linear homotopy theory for codifferential operator, leading to a direct sum decomposition of differential forms.
result Shows a new way to solve exterior differential systems, applicable to fundamental physics equations.

Develops experimental design for discovering missing physics in bioreactors.

problem Discovering missing physics in incomplete model structures of process systems.
method Combines universal differential equations and symbolic regression with sequential experimental design.
result Successfully recovered true model structure of a bioreactor using machine learning techniques.

DPC uses physics and neural nets to solve SDEs.

problem Solving stochastic differential equations with missing physics.
method Physics-data fusion with conditional maximum mean discrepancy (CMMD) loss.
result DPC achieves highly accurate solutions on benchmark examples.

Physics-informed DeepONets solve PDEs without paired data, predicting solutions quickly.

problem Lack of paired input-output data for solving PDEs.
method Physics-informed DeepONets use automatic differentiation to enforce physical laws as soft penalty constraints.
result Physics-informed DeepONets can solve PDEs without paired data, predicting solutions up to 3 orders of magnitude faster.

This study compares different thermodynamic structure-informed neural networks for solving differential equations.

problem Improving the accuracy and physical consistency of neural network solutions to differential equations.
method Comprehensive evaluation of various thermodynamic formulations in physics-informed neural networks.
result Newtonian-residual-based PINNs fail to reliably recover physical quantities, while structure-preserving formulations enhance accuracy and robustness.

While physics conveys knowledge of nature built from an interplay between observations and theory, it has been considered less importantly in deep neural networks. Especially, there are few works leveraging physics behaviors when the knowledge is given less explicitly. In this work, we propose a novel architecture call…

2019-02-08abs ↗pdf ↗

We present DiffTaichi, a new differentiable programming language tailored for building high-performance differentiable physical simulators. Based on an imperative programming language, DiffTaichi generates gradients of simulation steps using source code transformations that preserve arithmetic intensity and parallelism…

2019-10-01abs ↗pdf ↗

Physics-informed kernel learning integrates physical priors into machine learning models.

problem Tackles the integration of physical laws into machine learning models for improved accuracy and efficiency.
method Uses Fourier methods to approximate the kernel and minimizes a physics-informed risk function.
result Demonstrates PIKL outperforms physics-informed neural networks and traditional PDE solvers in various scenarios.

Proposes PI-VAE for solving SDEs with limited measurements.

problem Solving SDEs with limited measurements of system parameters.
method Physics-informed Variational Autoencoder (PI-VAE) integrating VAE and governing equations.
result Satisfactory accuracy and efficiency compared to PI-WGAN.

New method uses PINNs to efficiently compute Gerber-Shiu functions.

problem Calculating the Gerber-Shiu function efficiently.
method Physics-informed neural networks (PINNs) embedded with differential equations.
result Demonstrates good performance in approximating Gerber-Shiu functions.

Paper improves uncertainty quantification in PINNs using error bounds and solution bundles.

problem Uncertainty quantification in PINNs for differential equation systems.
method Two-step procedure with Bayesian Neural Networks and heteroscedastic variance.
result Improved uncertainty estimation over PINNs solutions in differential equation systems.

Paper discovers differential equations from data using neural networks and Bayesian methods.

problem Discovering differential equations from datasets using machine learning.
method Integrates neural network-based surrogates with Sparse Bayesian Learning (SBL).
result Proposes a robust model discovery algorithm and a Physics Informed Normalizing Flow (PINF).

Paper presents MF-PIDNN for physics-informed deep learning with low-fidelity data.

problem Challenges in systems with unknown or approximate governing differential equations and limited high-fidelity data.
method Transfer learning between physics-informed and data-driven deep learning models.
result Model provides accurate predictions even in data-scarce regions.

This work discovers governing equations from limited data using physics-informed deep learning.

problem Discovering governing equations from scarce and noisy data for complex systems.
method Physics-informed deep learning framework integrating neural networks, physics embedding, and sparse regression.
result The method effectively identifies governing equations from various spatiotemporal systems with different levels of data scarcity and noise.

In this paper we form a general conservation law that unifies a class of physics field theories. For this we first introduce the notion of a general field as a formal sum differential forms on a Minkowski manifold. Thereafter, we employ the action principle to define the conservation law for such general fields. By con…

2019-08-28abs ↗pdf ↗

Physics Informed Deep Kernel Learning improves prediction accuracy and uncertainty quantification.

problem Limited performance of deep kernel learning due to scarce or insufficient data.
method Integrates physics knowledge represented by differential equations with latent sources into deep kernel learning.
result Advantages in prediction accuracy and uncertainty quantification on synthetic and real-world datasets.

Intelligent agents need a physical understanding of the world to predict the impact of their actions in the future. While learning-based models of the environment dynamics have contributed to significant improvements in sample efficiency compared to model-free reinforcement learning algorithms, they typically fail to g…

2019-05-26abs ↗pdf ↗

Paper proposes a method to verify PINN fidelity using Fisher information from dynamical systems.

problem Quantifying PINN fidelity beyond simple trajectory prediction.
method Employing Fisher information for differentiable dynamical systems to compare PINN's learned equations with analytical models.
result PINN fidelity is verified by matching Fisher information landscapes of learned equations and analytical models.

A new hybrid approach combines physics and machine learning for porous media transport.

problem Simulating 2-phase immiscible transport in porous media.
method Physics-informed deep learning with adversarial neural networks and automatic differentiation.
result The model accurately simulates shock and rarefaction phenomena with limited data.

Framework augments physical models with deep learning for complex dynamics forecasting.

problem Forecasting complex dynamical phenomena with partial knowledge.
method APHYNITY framework: decomposes dynamics into physical and data-driven components.
result Framework accurately forecasts system evolution and identifies relevant parameters.

We consider differentiable maps in the setting of Abstract Differential Geometry and we study the conditions that ensure the uniqueness of differentials in this setting. In particular, we prove that smooth maps between smooth manifolds admit a unique differential, coinciding with the usual one. Thus smooth manifolds fo…

2013-11-25abs ↗pdf ↗

New method learns from non-uniform data and partial physical knowledge.

problem Identifying dynamical systems from non-uniformly sampled data.
method Physics-informed neural networks integrating numerical integration methods.
result Learning unknown kinetic rates and estimating parameters from non-uniform data.

Repulsive ensembles improve uncertainty estimates in PINNs for differential equations.

problem Improving uncertainty estimates in PINNs for differential equations.
method Employing repulsive ensembles (RE-PINN) with a repulsive term in the loss function.
result Repulsive ensembles produce more accurate uncertainty estimates and higher sample diversity.

Develops ΦΦ-DVAE for assimilating unstructured data into physical models.

problem Challenges in incorporating unstructured data into physical models.
method Physics-informed dynamical variational autoencoder (ΦΦ-DVAE) combining latent state-space model and VAE.
result Demonstrates data-efficient dynamics encoding with competitive performance and uncertainty quantification.