arXiv subjects

Ilse C. F. Ipsen

Publications and source records attributed to Ilse C. F. Ipsen.

At least 19 recordsLinked to original sources

arXiv2026

Revisiting column subset selection through the lens of submodularity

The problem is to select k columns with maximal volume from a real mxn matrix X. We show that the logarithm of the volume is a set submodular function on columns of X, and for full column-rank matrices X with sufficiently large singular values, it is a non-negative non-decreasing function. As a consequence, traditional Businger-Golub QR with column pivoting is a greedy algorithm, with a relative error of at most 37 percent. In contrast, Gu-Eisenstat strong rank-revealing QR is a 1-interchange algorithm, with a relative error of at most 50 percent. The higher accuracy, under this metric, of the simple QR with column pivoting confirms its well known effectiveness in practice. For general, possibly rank-deficient matrices, we derive probabilistic bounds for the absolute error based on a smoothed analysis. The above analyses are extended to finding kxk submatrices of maximal volume in symmetric positive-definite matrices.

math.NA↗

arXiv2026

Numerically Stable Cholesky-QR on GPU via Mixed-Precision Randomized Preconditioning

Cholesky-QR is among the fastest algorithms for computing the thin QR factorization of tall-and-skinny matrices on GPUs, relying entirely on BLAS-3 operations. However, it is numerically unstable: forming the Gram matrix squares the condition number, causing breakdown when $κ_2(\boldsymbol{A}) \gtrsim 10^8$. We present MRCQR (Mixed-Precision Randomized Cholesky-QR), a stable GPU algorithm that addresses this limitation. MRCQR uses a subsampled randomized trigonometric transform to construct a preconditioner $\boldsymbol{R}_s$ that reduces $κ_2(\boldsymbol{A}\boldsymbol{R}_s^{-1})$ to near unity with high probability, then applies Cholesky-QR in double precision to the preconditioned matrix. The key insight -- supported by perturbation analysis -- is that the preconditioner requires far less accuracy than the final result: single (FP32) precision suffices when $κ_2(\boldsymbol{A}) \lesssim 10^8$, and half (FP16) when $κ_2(\boldsymbol{A}) \lesssim 10^4$. MRCQR produces an explicit orthogonal factor $\widehat{\boldsymbol{Q}}$ satisfying $\|\boldsymbol{I} - \widehat{\boldsymbol{Q}}^\top\widehat{\boldsymbol{Q}}\|_2 = \cal O(\mathbf{u})$ ($\mathbf{u} \approx 10^{-16}$, double-precision unit roundoff) for condition numbers up to $10^{16}$, far beyond the $10^8$ limit of CholQR2. Experiments on an NVIDIA H100 GPU show that MRCQR (FP16) outperforms rand-cholQR by $1.4$--$1.8\times$ across all tested column counts and is $1.8$--$13.5\times$ faster than cuSOLVER geqrf, while the FP16 sketch (used when $κ_2(\boldsymbol{A}) \lesssim 10^4$) is $2\times$ cheaper than FP64 at no accuracy cost.

math.NA↗

arXiv2026

THE ROLE OF FOURIER ANALYSIS IN TWO DIMENSIONAL TOMOGRAPHY

We highlight the important role of the Fourier transform in deriving inversion formulas for the integral transforms of tomographic imaging. We demonstrate this principle by deriving inversion formulas for the divergent beam transform and the V-line transform, the latter arising in contemporary models of single-scattering optical tomography.

physics.optics↗

arXiv2026

Many (most?) column subset selection criteria are NP hard for a few columns

We consider a variety of criteria for selecting k representative columns from a real mxn matrix A, when sufficiently few columns are required, i.e., 1<= k<= min{rank(A), m/3}. The criteria include the following optimization problems: absolute volume and S-optimality maximization; norm, pseudo-inverse norm, and condition minimization number in the two-norm, Frobenius norm and Schatten p-norms for p>2; stable rank maximization; and the new criterion of relative volume maximization, which is inversely proportional to a power of the condition number. We show that these criteria are NP hard and many do not admit polynomial time approximation schemes (PTAS). To formulate the optimization problems as decision problems, we derive optimal values for the subset selection criteria, as well as expressions for partitioned pseudo-inverses. The results for minimization of the pseudo-inverse in the Frobenius norm are applicable to trace optimization in A-optimal design.

math.NA↗

arXiv2026

Perturbation Analysis for Preconditioned Normal Equations in Mixed Precision

For real matrices of full column-rank, we analyze the conditioning of several types of normal equations that are preconditioned by a randomized preconditioner computed in lower precision. These include symmetrically preconditioned normal equations, half-preconditioned normal equations, seminormal equations and not-normal equations. Our perturbation bounds are realistic and informative, and suggest that the conditioning depends only mildly on the quality of the preconditioner; however, it does depend on the size of the least squares residual -- even if the normal equations do not originate from a least squares problem. We illustrate that a randomized preconditioner can deliver a solution accuracy comparable to that of Matlab's mldivide command, is efficient in practice, and well-suited to GPU implementations. For the computation of the preconditioner, we propose an automatic selection of the precision, based on a fast condition number estimation in lower precision.

math.NA↗

arXiv2025

Solution of Least Squares Problems with Randomized Preconditioned Normal Equations

We consider the solution of full column-rank least squares problems by means of normal equations that are preconditioned, symmetrically or non-symmetrically, with a randomized preconditioner. With an effective preconditioner, the solutions from the preconditioned normal equations are almost as accurate as those from the QR-based Matlab backslash (mldivide) command -- even for highly illconditioned matrices. This means the accuracy of the preconditioned normal equations depends on the residual of the original least squares problem. We present non-intuitive but realistic perturbation bounds for the relative error in the computed solutions and show that, with an effective preconditioner, these bounds are essentially equal to the perturbation bound for the original least squares problem. Probabilitistic condition number bounds corroborate the effectiveness of the randomized preconditioner computed with small amounts of sampling.

math.NA↗

arXiv2025

Stochastic Trace Optimization of Parameter Dependent Matrices Based on Statistical Learning Theory

We consider matrices $\boldsymbol{A}(\boldsymbolθ)\in\mathbb{R}^{m\times m}$ that depend, possibly nonlinearly, on a parameter $\boldsymbolθ$ from a compact parameter space $Θ$. We present a Monte Carlo estimator for minimizing $\text{trace}(\boldsymbol{A}(\boldsymbolθ))$ over all $\boldsymbolθ\inΘ$, and determine the sampling amount so that the backward error of the estimator is bounded with high probability. We derive two types of bounds, based on epsilon nets and on generic chaining. Both types predict a small sampling amount for matrices $\boldsymbol{A}(\boldsymbolθ)$ with small offdiagonal mass, and parameter spaces $Θ$ of small ``size.'' Dependence on the matrix dimension~$m$ is only weak or not explicit. The bounds based on epsilon nets are easier to evaluate and come with fully specified constants. In contrast, the bounds based on chaining depend on the Talagrand functionals which are difficult to evaluate, except in very special cases. Comparisons between the two types of bounds are difficult, although the literature suggests that chaining bounds can be superior.

stat.ML↗

arXiv2024

Stable Rank and Intrinsic Dimension of Real and Complex Matrices

The notion of `stable rank' of a matrix is central to the analysis of randomized matrix algorithms, covariance estimation, deep neural networks, and recommender systems. We compare the properties of the stable rank and intrinsic dimension of real and complex matrices to those of the classical rank. Basic proofs and examples illustrate that the stable rank does not satisfy any of the fundamental rank properties, while the intrinsic dimension satisfies a few. In particular, the stable rank and intrinsic dimension of a submatrix can exceed those of the original matrix; adding a Hermitian positive semi-definite matrix can lower the intrinsic dimension of the sum; and multiplication by a nonsingular matrix can drastically change the stable rank and the intrinsic dimension. We generalize the concept of stable rank to the p-stable in any Schatten p-norm, thereby unifying the concepts of stable rank and intrinsic dimension: The stable rank is the 2-stable rank, while the intrinsic dimension is the 1-stable rank of a Hermitian positive semi-definite matrix. We derive sum and product inequalities for the pth root of the p-stable rank, and show that it is well-conditioned in the norm-wise absolute sense. The conditioning improves if the matrix and the perturbation are Hermitian positive semi-definite.

math.NA↗

arXiv2024

Stochastic Rounding Implicitly Regularizes Tall-and-Thin Matrices

Motivated by the popularity of stochastic rounding in the context of machine learning and the training of large-scale deep neural network models, we consider stochastic nearness rounding of real matrices $\mathbf{A}$ with many more rows than columns. We provide novel theoretical evidence, supported by extensive experimental evaluation that, with high probability, the smallest singular value of a stochastically rounded matrix is well bounded away from zero -- regardless of how close $\mathbf{A}$ is to being rank deficient and even if $\mathbf{A}$ is rank-deficient. In other words, stochastic rounding \textit{implicitly regularizes} tall and skinny matrices $\mathbf{A}$ so that the rounded version has full column rank. Our proofs leverage powerful results in random matrix theory, and the idea that stochastic rounding errors do not concentrate in low-dimensional column spaces.

cs.LG↗

arXiv2024

Stochastic Rounding 2.0, with a View towards Complexity Analysis

Stochastic Rounding is a probabilistic rounding mode that is surprisingly effective in large-scale computations and low-precision arithmetic. Its random nature promotes error cancellation rather than error accumulation, resulting in slower growth of roundoff errors as the problem size increases, especially when compared to traditional deterministic rounding methods, such as rounding-to-nearest. We advocate for SR as a foundational tool in the complexity analysis of algorithms, and suggest several research directions.

math.NA↗

arXiv2024

A randomized preconditioned Cholesky-QR algorithm

We a present and analyze rpCholesky-QR, a randomized preconditioned Cholesky-QR algorithm for computing the thin QR factorization of real mxn matrices with rank n. rpCholesky-QR has a low orthogonalization error, a residual on the order of machine precision, and does not break down for highly singular matrices. We derive rigorous and interpretable two-norm perturbation bounds for rpCholesky-QR that require a minimum of assumptions. Numerical experiments corroborate the accuracy of rpCholesky-QR for preconditioners sampled from as few as 3n rows, and illustrate that the two-norm deviation from orthonormality increases with only the condition number of the preconditioned matrix, rather than its square -- even if the original matrix is numerically singular.

math.NA↗

arXiv2024

Small singular values can increase in lower precision

We perturb a real matrix $A$ of full column rank, and derive lower bounds for the smallest singular values of the perturbed matrix, in terms of normwise absolute perturbations. Our bounds, which extend existing lower-order expressions, demonstrate the potential increase in the smallest singular values, and represent a qualitative model for the increase in the small singular values after a matrix has been downcast to a lower arithmetic precision. Numerical experiments confirm the qualitative validity of this model and its ability to predict singular values changes in the presence of decreased arithmetic precision.

math.NA↗

arXiv2022

BayesCG As An Uncertainty Aware Version of CG

The Bayesian Conjugate Gradient method (BayesCG) is a probabilistic generalization of the Conjugate Gradient method (CG) for solving linear systems with real symmetric positive definite coefficient matrices. Our CG-based implementation of BayesCG under a structure-exploiting prior distribution represents an 'uncertainty-aware' version of CG. Its output consists of CG iterates and posterior covariances that can be propagated to subsequent computations. The covariances have low-rank and are maintained in factored form. This allows easy generation of accurate samples to probe uncertainty in downstream computations. Numerical experiments confirm the effectiveness of the low-rank posterior covariances.

math.NA↗

arXiv2022

Statistical Properties of the Probabilistic Numeric Linear Solver BayesCG

We analyse the calibration of BayesCG under the Krylov prior, a probabilistic numeric extension of the Conjugate Gradient (CG) method for solving systems of linear equations with symmetric positive definite coefficient matrix. Calibration refers to the statistical quality of the posterior covariances produced by a solver. Since BayesCG is not calibrated in the strict existing notion, we propose instead two test statistics that are necessary but not sufficient for calibration: the Z-statistic and the new S-statistic. We show analytically and experimentally that under low-rank approximate Krylov posteriors, BayesCG exhibits desirable properties of a calibrated solver, is only slightly optimistic, and is computationally competitive with CG.

math.NA↗

arXiv2022

Robust Parameter Identifiability Analysis via Column Subset Selection

We advocate a numerically reliable and accurate approach for practical parameter identifiability analysis: Applying column subset selection (CSS) to the sensitivity matrix, instead of computing an eigenvalue decomposition of the Fischer information matrix. Identifiability analysis via CSS has three advantages: (i) It quantifies reliability of the subsets of parameters selected as identifiable and unidentifiable. (ii) It establishes criteria for comparing the accuracy of different algorithms. (iii) The implementations are numerically more accurate and reliable than eigenvalue methods applied to the Fischer matrix, yet without an increase in computational cost. The effectiveness of the CSS methods is illustrated with extensive numerical experiments on sensitivity matrices from six physical models, as well as on adversarial synthetic matrices. Among the CSS methods, we recommend an implementation based on the strong rank-revealing QR algorithm because of its rigorous accuracy guarantees for both identifiable and non-identifiable parameters.

math.NA↗

arXiv2022

Precision-aware Deterministic and Probabilistic Error Bounds for Floating Point Summation

We analyze the forward error in the floating point summation of real numbers, for computations in low precision or extreme-scale problem dimensions that push the limits of the precision. We present a systematic recurrence for a martingale on a computational tree, which leads to explicit and interpretable bounds without asymptotic big-O terms. Two probability parameters strengthen the precision-awareness of our bounds: one parameter controls the first order terms in the summation error, while the second one is designed for controlling higher order terms in low precision or extreme-scale problem dimensions. Our systematic approach yields new deterministic and probabilistic error bounds for three classes of mono-precision algorithms: general summation, shifted general summation, and compensated (sequential) summation. Extension of our systematic error analysis to mixed-precision summation algorithms that allow any number of precisions yields the first probabilistic bounds for the mixed-precision FABsum algorithm. Numerical experiments illustrate that the probabilistic bounds are accurate, and that among the three classes of mono-precision algorithms, compensated summation is generally the most accurate. As for mixed precision algorithms, our recommendation is to minimize the magnitude of intermediate partial sums relative to the precision in which they are computed.

math.NA↗

arXiv2022

Monte Carlo Methods for Estimating the Diagonal of a Real Symmetric Matrix

For real symmetric matrices that are accessible only through matrix vector products, we present Monte Carlo estimators for computing the diagonal elements. Our probabilistic bounds for normwise absolute and relative errors apply to Monte Carlo estimators based on random Rademacher, sparse Rademacher, normalized and unnormalized Gaussian vectors, and to vectors with bounded fourth moments. The novel use of matrix concentration inequalities in our proofs represents a systematic model for future analyses. Our bounds mostly do not depend on the matrix dimension, target different error measures than existing work, and imply that the accuracy of the estimators increases with the diagonal dominance of the matrix. An application to derivative-based global sensitivity metrics corroborates this, as do numerical experiments on synthetic test matrices. We recommend against the use in practice of sparse Rademacher vectors, which are the basis for many randomized sketching and sampling algorithms, because they tend to deliver barely a digit of accuracy even under large sampling amounts.

math.NA↗

arXiv2021

Deterministic and Probabilistic Error Bounds for Floating Point Summation Algorithms

We analyse the forward error in the floating point summation of real numbers, from algorithms that do not require recourse to higher precision or better hardware. We derive informative explicit expressions, and new deterministic and probabilistic bounds for errors in three classes of algorithms: general summation,shifted general summation, and compensated (sequential) summation. Our probabilistic bounds for general and shifted general summation hold to all orders. For compensated summation, we also present deterministic and probabilistic first and second order bounds, with a first order bound that differs from existing ones. Numerical experiments illustrate that the bounds are informative and that among the three algorithm classes, compensated summation is generally the most accurate method.

math.NA↗