arXiv research

Subsampled Randomized Hadamard Transform (SRHT), a popular random projection method that can efficiently project a $d$-dimensional data into $r$-dimensional space ($r \ll d$) in $O(dlog(d))$ time, has been widely used to address the challenge of high-dimensionality in machine learning. SRHT works by rotating the input …

Paper computes link determinants using Fourier-Hadamard transforms.

Uniform approximations for RHTs improve kernel approximation and distance estimation.

Unified methodology for statistical inference in least squares and PCA via randomized sketching.

A well-known problem in data science and machine learning is {\em linear regression}, which is recently extended to dynamic graphs. Existing exact algorithms for updating the solution of dynamic graph regression require at least a linear time (in terms of $n$: the size of the graph). However, this time complexity might…

A new class of harmonic Hadamard manifolds, those spaces called of hypergeometric type, is defined in terms of Gauss hypergeometric equations. Spherical Fourier transform defined on a harmonic Hadamard manifold of hypergeometric type admits an inversion formula. A characterization of harmonic Hadamard manifold being of…

DFRot improves LLMs by reducing outlier and massive activation effects.

We prove two injectivity theorems for the geodesic ray transform on two-dimensional, complete, simply connected Riemannian manifolds with non-positive Gaussian curvature, also known as Cartan-Hadamard manifolds. The first theorem is concerned with bounded non-positive curvature and the second with decaying non-positive…

New insights into identifying mixtures of product distributions using Hadamard extensions.

Study fourth order Schrödinger equation on Cartan-Hadamard manifolds, proving existence, scattering, and blow-up results.

A new iterative low complexity algorithm has been presented for computing the Walsh-Hadamard transform (WHT) of an $N$ dimensional signal with a $K$-sparse WHT, where $N$ is a power of two and $K = O(N^α)$, scales sub-linearly in $N$ for some $0 < α< 1$. Assuming a random support model for the non-zero transform domain…

We study the geodesic X-ray transform on Cartan-Hadamard manifolds, and prove solenoidal injectivity of this transform acting on functions and tensor fields of any order. The functions are assumed to be exponentially decaying if the sectional curvature is bounded, and polynomially decaying if the sectional curvature de…

Geometrically proves majorizing measure theorem on Hadamard manifolds.

Variational inference offers scalable and flexible tools to tackle intractable Bayesian inference of modern statistical models like Bayesian neural networks and Gaussian processes. For largely over-parameterized models, however, the over-regularization property of the variational objective makes the application of vari…

The Weyl transform is introduced as a rich framework for data representation. Transform coefficients are connected to the Walsh-Hadamard transform of multiscale autocorrelations, and different forms of dyadic periodicity in a signal are shown to appear as different features in its Weyl coefficients. The Weyl transform …

Extends DCP framework to Hadamard manifolds for geodesically convex functions.

Corrects bias in random sampling matrices for improved ML methods.

Random projection (RP) is a classical technique for reducing storage and computational costs. We analyze RP-based approximations of convex programs, in which the original optimization problem is approximated by the solution of a lower-dimensional problem. Such dimensionality reduction is essential in computation-limite…

Linear mixed models (LMMs) are used extensively to model dependecies of observations in linear regression and are used extensively in many application areas. Parameter estimation for LMMs can be computationally prohibitive on big data. State-of-the-art learning algorithms require computational complexity which depends …

We consider a least squares regression problem where the data has been generated from a linear model, and we are interested to learn the unknown regression parameters. We consider "sketch-and-solve" methods that randomly project the data first, and do regression after. Previous works have analyzed the statistical and c…

Sketching, a dimensionality reduction technique, has received much attention in the statistics community. In this paper, we study sketching in the context of Newton's method for solving finite-sum optimization problems in which the number of variables and data points are both large. We study two forms of sketching that…

Study spherical Fourier transform on hypergeometric type harmonic manifolds.

Over-parameterized models, such as DeepNets and ConvNets, form a class of models that are routinely adopted in a wide variety of applications, and for which Bayesian inference is desirable but extremely challenging. Variational inference offers the tools to tackle this challenge in a scalable way and with some degree o…

The paper derives expansions for Green's operators and resolvents using Hadamard methods.

Harmonic maps between pinched Hadamard surfaces are quasi-conformal.

We give a new algorithm for approximating the Discrete Fourier transform of an approximately sparse signal that has been corrupted by worst-case $L_0$ noise, namely a bounded number of coordinates of the signal have been corrupted arbitrarily. Our techniques generalize to a wide range of linear transformations that are…

Geometrically constructs twist-field correlation functions in CFT.

Cyclic projections in Hadamard spaces can be irregular, unlike in Hilbert spaces.

Study shows spectrum properties for specific Hadamard manifolds.

Sharp curvature inequality extends Euclidean isoperimetric inequality to Cartan-Hadamard manifolds.

The paper provides formulas for Hadamard coefficients using Green's operators.

New algorithm reduces sketching dimension to effective problem size.

We propose unitary group convolutions (UGConvs), a building block for CNNs which compose a group convolution with unitary transforms in feature space to learn a richer set of representations than group convolution alone. UGConvs generalize two disparate ideas in CNN architecture, channel shuffling (i.e. ShuffleNet) and…

Formula for Hadamard coefficients from Green's operators on spacetimes.

Upper bounds for circumradius in Hadamard surfaces with curvature constraints.

Paper proposes a method to speed up DNNs by quantizing Winograd/Toom-Cook convolutions.

New theory approximates functions between metric spaces using random probability measures.

Solves Plateau problem for surfaces in pinched curvature manifolds.

Proves Hadamard states for Dirac fields on manifolds with timelike boundaries.

Various results based on some convexity assumptions (involving the exponential map along with affine maps, geodesics and convex hulls) have been recently established on Hadamard manifolds. In this paper we prove that these conditions are mutually equivalent and they hold if and only if the Hadamard manifold is isometri…

New theorem extends Hadamard's to transversely affine geometry.

The paper proves Liouville-type theorems on Hadamard manifolds.

Maps between Hadamard manifolds are quasi-isometric to harmonic maps.

Given a closed submanifold, or a compact regular domain, in euclidean space, we consider the Riesz energy defined as the double integral of some power of the distance between pairs of points. When this integral diverges, we compare two different regularization techniques (Hadamard's finite part and analytic continuatio…

The study extends isoperimetric inequalities to non-positive curvature spaces.

Efficient algorithm for Hadamard decomposition of matrices.

Study KKT conditions for multi-objective optimization on Hadamard manifolds.

We use the combinatorial harmonic map theory to study the isometric actions of discrete groups on Hadamard spaces. Given a finitely generated group acting by automorphisms, properly discontinuously and cofinitely on a simplicial complex and its isometric action on a Hadamard space, we formulate criterions for the actio…