arXiv research

New method for causal inference with observed covariates improves learning rates.

This monograph develops the theory of covariant Schrödinger semigroups acting on sections of vector bundles over noncompact Riemannian manifolds from scratch. Contents: I. Sobolev spaces on vector bundles II. Smooth heat kernels on vector bundles III. Basis differential operators in Riemannian manifolds IV. Some specif…

Kalman filtering and smoothing algorithms are used in many areas, including tracking and navigation, medical applications, and financial trend filtering. One of the basic assumptions required to apply the Kalman smoothing framework is that error covariance matrices are known and given. In this paper, we study a general…

New algorithm improves matrix estimation with one-sided covariates.

Bayesian methods estimate regression functions on submanifolds using graph Laplacian eigenbasis.

Study nonparametric covariance function estimation for noisy data.

Estimates covariance matrices for matrix-variate data via core covariance geometry.

Optimizes spectral density estimation for stationary and nonstationary processes.

In modeling multivariate time series, it is important to allow time-varying smoothness in the mean and covariance process. In particular, there may be certain time intervals exhibiting rapid changes and others in which changes are slow. If such time-varying smoothness is not accounted for, one can obtain misleading inf…

GLSKF improves tensor completion by capturing both global and local variations.

New GP kernels avoid mean reversion without losing smoothness.

Study on Blaschke locus with covariance metric properties.

On a Riemannian or a semi-Riemannian manifold, the metric determines invariants like the Levi-Civita connection and the Riemann curvature. If the metric becomes degenerate (as in singular semi-Riemannian geometry), these constructions no longer work, because they are based on the inverse of the metric, and on related o…

Study optimizes decisions in real-time using inexact simulation solutions.

We review recent probabilistic results on covariant Schrödinger operators on vector bundles over (possibly locally infinite) weighted graphs, and explain applications like semiclassical limits. We also clarify the relationship between these results and their formal analogues on smooth (possibly noncompact) Riemannian m…

Proposes a new ridge estimator for smooth covariates with adaptive centering.

Improved estimators for causal inference using cross-fitting and undersmoothing.

Method regularizes Cholesky factors to detect nonstationarity in longitudinal data.

Paper introduces a novel method for dynamic covariance estimation with random forests.

We study a non-parametric multi-armed bandit problem with stochastic covariates, where a key complexity driver is the smoothness of payoff functions with respect to covariates. Previous studies have focused on deriving minimax-optimal algorithms in cases where it is a priori known how smooth the payoff functions are. I…

Interactive privacy mechanisms improve spectral density estimation under local differential privacy.

Estimates personalized treatment response curves using covariates.

New method provides valid confidence intervals for spatial associations.

Paper addresses covariate shift in deep learning regression models.

Designing a covariance function that represents the underlying correlation is a crucial step in modeling complex natural systems, such as climate models. Geospatial datasets at a global scale usually suffer from non-stationarity and non-uniformly smooth spatial boundaries. A Gaussian process regression using a non-stat…

The n-th order covariant derivative on a smooth manifold with an affine connection is a differential operator which turns a function into a tensor field of type (0,n). In this paper the properties of this operatior related to the permutation of indices are investigated by means of non-associative algebra. The general f…

ULA estimates covariance of log-concave distributions efficiently.

Deep model predicts shapes of curves with multiple covariates.

We present a general probabilistic perspective on Gaussian filtering and smoothing. This allows us to show that common approaches to Gaussian filtering/smoothing can be distinguished solely by their methods of computing/approximating the means and covariances of joint probabilities. This implies that novel filters and …

Consider a smooth manifold with a smooth metric which changes bilinear type from Riemann to Lorentz on a hypersurface $Σ$ with radical tangent to $Σ$. Two natural bilinear symmetric forms appear there, and we use it to analyze the geometry of $Σ$. We show the way in which these forms control the smooth extensibility ov…

Paper estimates non-causal graphical models using covariance extension and transportation distance.

Survey on smooth function and form density in Riemannian Sobolev spaces.

Many popular statistical models, such as factor and random effects models, give arise a certain type of covariance structures that is a summation of low rank and sparse matrices. This paper introduces a penalized approximation framework to recover such model structures from large covariance matrix estimation. We propos…

Develops a novel ML smoothing method for incomplete data in state-space models.

Improves Gaussian process models for large datasets.

Given $n$ i.i.d. observations of a random vector $(X,Z)$, where $X$ is a high-dimensional vector and $Z$ is a low-dimensional index variable, we study the problem of estimating the conditional inverse covariance matrix $Ω(z) = (E[(X-E[X \mid Z])(X-E[X \mid Z])^T \mid Z=z])^{-1}$ under the assumption that the set of non…

SPARKLE handles high-dimensional covariates for online decision-making.

In this paper we consider the use of the space vs. time Kronecker product decomposition in the estimation of covariance matrices for spatio-temporal data. This decomposition imposes lower dimensional structure on the estimated covariance matrix, thus reducing the number of samples required for estimation. To allow a sm…

A flag area measure on an $n$-dimensional euclidean vector space is a continuous translation-invariant valuation with values in the space of signed measures on the flag manifold consisting of a unit vector $v$ and a $(p+1)$-dimensional linear subspace containing $v$ with $0 \leq p \leq n-1$. Using local parallel sets, …

We propose a flexible nonparametric regression method for ultrahigh-dimensional data. As a first step, we propose a fast screening method based on the favored smoothing bandwidth of the marginal local constant regression. Then, an iterative procedure is developed to recover both the important covariates and the regress…

For a smooth manifold $M$, it was shown in \cite{BPH} that every affine connection on the tangent bundle $TM$ naturally gives rise to covariant differentiation of multivector fields (MVFs) and differential forms along MVFs. In this paper, we generalize the covariant derivative of \cite{BPH} and construct covariant deri…

The paper develops methods to reduce deployment risk under dynamic covariate shifts.

The paper introduces Shapley curves for measuring variable importance in nonparametric settings.

Adaptive Bayesian model for covariate-dependent power spectra analysis.

Bayesian model clusters brain activity time series.

We study differential cohomology on categories of globally hyperbolic Lorentzian manifolds. The Lorentzian metric allows us to define a natural transformation whose kernel generalizes Maxwell's equations and fits into a restriction of the fundamental exact sequences of differential cohomology. We consider smooth Pontry…

Geometric cohomology model uses co-oriented maps to define a product structure.

Paper develops an online covariance estimator for nonsmooth stochastic approximation problems.