arXiv research

Bayesian inference for stochastic differential equations using Wishart diffusions.

Estimates neural drift for stochastic equations, improving inference on noisy data.

New framework discovers PDEs from sparse, noisy data.

Study uses neural networks to solve complex equations efficiently.

X-TFC solves parametric DEs with neural networks and physics constraints.

We derive upper bounds on the complexity of ReLU neural networks approximating the solution maps of parametric partial differential equations. In particular, without any knowledge of its concrete shape, we use the inherent low-dimensionality of the solution manifold to obtain approximation rates which are significantly…

One goal in Bayesian machine learning is to encode prior knowledge into prior distributions, to model data efficiently. We consider prior knowledge from systems of linear partial differential equations together with their boundary conditions. We construct multi-output Gaussian process priors with realizations in the so…

Many equations of mathematical physics are described by differential polynomials, that is by polynomials in the derivatives of a certain number of functions. However, up to the knowledge of the author, differential algebra in a modern setting has never been applied to study the specific algebraic feature of such equati…

This work leverages recent advances in probabilistic machine learning to discover conservation laws expressed by parametric linear equations. Such equations involve, but are not limited to, ordinary and partial differential, integro-differential, and fractional order operators. Here, Gaussian process priors are modifie…

Logic approach finds real singularities in differential equations.

Deep neural network approximates multivariate option pricing.

We propose definitions of homogeneity and projective equivalence for systems of ordinary differential equations of order greater than two, which allow us to generalize the concept of a spray (for systems of order two). We show that the Euler-Lagrange fields of parametric Lagrangians of order greater than one which are …

New minimal surfaces in 4D space derived from parametric equations.

We introduce a technique based on the singular vector canonical correlation analysis (SVCCA) for measuring the generality of neural network layers across a continuously-parametrized set of tasks. We illustrate this method by studying generality in neural networks trained to solve parametrized boundary value problems ba…

This note introduces a regression technique for finding a class of nonlinear integro-differential operators from data. The method parametrizes the spatial operator with neural networks and Fourier transforms such that it can fit a class of nonlinear operators without needing a library of a priori selected operators. We…

Deep adaptive sampling improves surrogate modeling for complex systems.

We introduce a novel paradigm for learning non-parametric drift and diffusion functions for stochastic differential equation (SDE). The proposed model learns to simulate path distributions that match observations with non-uniform time increments and arbitrary sparseness, which is in contrast with gradient matching that…

We solve the mean parametrization of von Mises-Fisher distribution.

New ADANNs improve PDE approximations.

Framework improves data-driven ROMs for complex systems using Bayesian operator inference.

Deep learning solves high-dimensional PDEs efficiently.

Extends machine learning models for analytic boundary conditions in differential equations.

New model solves PDEs using probabilistic random grids.

Paper explores solving HJB equations using neural networks.

New method speeds up SDE inference by matching moments to FPK equation.

Model inference for dynamical systems aims to estimate the future behaviour of a system from observations. Purely model-free statistical methods, such as Artificial Neural Networks, tend to perform poorly for such tasks. They are therefore not well suited to many questions from applications, for example in Bayesian fil…

Meta-learning base distributions for efficient PDE solutions.

In this paper, position vectors of a time-like curve with respect to standard frame of Minkowski space E$^3_1$ are studied in terms of Frenet equations. First, we prove that position vector of every time-like space curve in Minkowski space E$^3_1$ satisfies a vector differential equation of fourth order. The general so…

Method learns SDEs from data snapshots.

The application of Stochastic Differential Equations (SDEs) to the analysis of temporal data has attracted increasing attention, due to their ability to describe complex dynamics with physically interpretable equations. In this paper, we introduce a non-parametric method for estimating the drift and diffusion terms of …

The aim of this paper is to construct and analyze solutions to a class of Hamilton-Jacobi-Bellman equations with range bounds on the optimal response variable. Using the Riccati transformation we derive and analyze a fully nonlinear parabolic partial differential equation for the optimal response function. We construct…

Enhanced autoencoders improve ROMs for PDEs by capturing essential properties.

We algorithmically construct multi-output Gaussian process priors which satisfy linear differential equations. Our approach attempts to parametrize all solutions of the equations using Gröbner bases. If successful, a push forward Gaussian process along the paramerization is the desired prior. We consider several exampl…

The main purpose of this paper is to revisit the well known potentials, called stress functions, needed in order to study the parametrizations of the stress equations, respectively provided by G.B. Airy (1863) for 2-dimensional elasticity, then by E. Beltrami (1892), J.C. Maxwell (1870) and G. Morera (1892) for 3-dimen…

Algorithm finds Liouvillian solutions for planar rational vector fields.

We propose kernel-based collocation methods for numerical solutions to Heath-Jarrow-Morton models with Musiela parametrization. The methods can be seen as the Euler-Maruyama approximation of some finite dimensional stochastic differential equations, and allow us to compute the derivative prices by the usual Monte Carlo…

New method combines ODE solvers with Bayesian inference for efficient model training.

Alternative construction of quasi-Fuchsian flows using vortex equations.

Consider the problem of learning the drift coefficient of a stochastic differential equation from a sample path. In this paper, we assume that the drift is parametrized by a high dimensional vector. We address the question of how long the system needs to be observed in order to learn this vector of parameters. We prove…

Solitons are special polygon midpoints under affine transformations.

New algorithms for approximating stochastic processes efficiently.

In this paper, position vector of a spacelike general helix with respect to standard frame in Minkowski space E$^3_1$ are studied in terms of Frenet equations. First, a vector differential equation of third order is constructed to determine position vector of an arbitrary spacelike general helix. In terms of solution, …

In this paper, we prove that the position vector of every space curve satisfies a vector differential equation of fourth order. Also, we determine the parametric representation of the position vector $ψ=\Big(ψ_1,ψ_2,ψ_3\Big)$ of general helices from the intrinsic equations $κ=κ(s)$ and $τ=τ(s)$ where $κ$ and $τ$ are th…

Study of curves in Lie sphere geometry using moving frames and variational principles.

Self-test loss functions improve data-driven modeling of weak-form operators and gradient flows.

DeepONet learns operators for PDEs with varying parameters and initial conditions.

The goal of this paper is to classify parametrically parabolic submanifolds in any codimension. First, we describe the ones that are ruled and show that they are the only parabolic submanifolds that admit an isometric immersion as a hypersurface. Then, we classify the nonruled ones by two different means. In fact, we p…

Reduces function approximation dimensions from high to low with sparse data.