Inversion-Free Natural Gradient Descent on Riemannian Manifolds

• URL: http://arxiv.org/abs/2604.02969v1
• Date: Fri, 03 Apr 2026 11:08:59 GMT
• Title: Inversion-Free Natural Gradient Descent on Riemannian Manifolds
• Abstract summary: This paper proposes an inversion-free natural gradient method for probability distributions with convex parameters.<n>The manifold setting offers several advantages: one can implicitly enforce parameter constraints.<n>A limited-memory variant of the algorithm with sub-quadratic storage complexity is proposed.
• Score: 5.556612138252781
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The natural gradient method is widely used in statistical optimization, but its standard formulation assumes a Euclidean parameter space. This paper proposes an inversion-free stochastic natural gradient method for probability distributions whose parameters lie on a Riemannian manifold. The manifold setting offers several advantages: one can implicitly enforce parameter constraints such as positive definiteness and orthogonality, ensure parameters are identifiable, or guarantee regularity properties of the objective like geodesic convexity. Building on an intrinsic formulation of the Fisher information matrix (FIM) on a manifold, our method maintains an online approximation of the inverse FIM, which is efficiently updated at quadratic cost using score vectors sampled at successive iterates. In the Riemannian setting, these score vectors belong to different tangent spaces and must be combined using transport operations. We prove almost-sure convergence rates of $O(\log{s}/s^α)$ for the squared distance to the minimizer when the step size exponent $α>2/3$. We also establish almost-sure rates for the approximate FIM, which now accumulates transport-based errors. A limited-memory variant of the algorithm with sub-quadratic storage complexity is proposed. Finally, we demonstrate the effectiveness of our method relative to its Euclidean counterparts on variational Bayes with Gaussian approximations and normalizing flows.

Related papers

• Optimality of Sub-network Laplace Approximations: New Results and Methods [0.0]
  Sub-network Laplace approximations construct surrogates by restricting attention to a small subset of parameters.<n>We show that sub-network Laplace methods systematically underestimate the predictive variance of the full Laplace posterior.<n>We propose two principled, analytically grounded sub-network Hessian approximations.
  arXiv  Detail & Related papers  (2026-05-09T17:37:06Z)
• Nonparametric Estimation of Isotropic Covariance Function [6.456477445670593]
  A nonparametric model is constructed to approximate arbitrary isotropic covariance functions valid in $mathbbRinfty$.<n>A sieve maximum likelihood (sML) estimation is then developed to estimate the unknown isotropic covaraince function valid in $mathbbRinfty$.<n>Our approach outperforms the parametric methods in reducing bias due to model misspecification and also the nonparametric methods in terms of having significantly lower values of expected $L_infty$ and $L$ norms.
  arXiv  Detail & Related papers  (2026-04-24T07:51:27Z)
• Variational Entropic Optimal Transport [67.76725267984578]
  We propose Variational Entropic Optimal Transport (VarEOT) for domain translation problems.<n>VarEOT is based on an exact variational reformulation of the log-partition $log mathbbE[exp(cdot)$ as a tractable generalization over an auxiliary positive normalizer.<n> Experiments on synthetic data and unpaired image-to-image translation demonstrate competitive or improved translation quality.
  arXiv  Detail & Related papers  (2026-02-02T15:48:44Z)
• Trust-Region Sequential Quadratic Programming for Stochastic Optimization with Random Models [57.52124921268249]
  We propose a Trust Sequential Quadratic Programming method to find both first and second-order stationary points. To converge to first-order stationary points, our method computes a gradient step in each iteration defined by minimizing a approximation of the objective subject. To converge to second-order stationary points, our method additionally computes an eigen step to explore the negative curvature the reduced Hessian matrix.
  arXiv  Detail & Related papers  (2024-09-24T04:39:47Z)
• Decomposed Global Optimization for Robust Point Matching with Low-Dimensional Branching [41.05165517541873]
  We introduce a novel global optimization method for align partially overlapping point sets.<n>Our method exhibits superior robustness to non-rigid deformations, positional noise and outliers.<n> Experiments on 2D and 3D synthetic and real-world data demonstrate that our method, compared to state-of-the-art approaches, exhibits superior robustness to outliers.
  arXiv  Detail & Related papers  (2024-05-14T13:28:57Z)
• Riemannian stochastic optimization methods avoid strict saddle points [68.80251170757647]
  We show that policies under study avoid strict saddle points / submanifolds with probability 1. This result provides an important sanity check as it shows that, almost always, the limit state of an algorithm can only be a local minimizer.
  arXiv  Detail & Related papers  (2023-11-04T11:12:24Z)
• Low-complexity subspace-descent over symmetric positive definite manifold [9.346050098365648]
  We develop low-complexity algorithms for the minimization of functions over the symmetric positive definite (SPD) manifold. The proposed approach utilizes carefully chosen subspaces that allow the update to be written as a product of the Cholesky factor of the iterate and a sparse matrix.
  arXiv  Detail & Related papers  (2023-05-03T11:11:46Z)
• Decentralized Riemannian natural gradient methods with Kronecker-product approximations [11.263837420265594]
  We present an efficient decentralized natural gradient descent (DRNGD) method for solving decentralized manifold optimization problems. By performing the communications over the Kronecker factors, a high-quality approximation of the RFIM can be obtained in a low cost.
  arXiv  Detail & Related papers  (2023-03-16T19:36:31Z)
• Kernel-based off-policy estimation without overlap: Instance optimality beyond semiparametric efficiency [53.90687548731265]
  We study optimal procedures for estimating a linear functional based on observational data. For any convex and symmetric function class $mathcalF$, we derive a non-asymptotic local minimax bound on the mean-squared error.
  arXiv  Detail & Related papers  (2023-01-16T02:57:37Z)
• MARS via LASSO [1.5199437137239338]
  We propose and study a natural variant of the MARS method. Our method is based on at least squares estimation over a convex class of functions. Our estimator can be computed via finite-dimensional convex optimization.
  arXiv  Detail & Related papers  (2021-11-23T07:30:33Z)
• Unified Convergence Analysis for Adaptive Optimization with Moving Average Estimator [75.05106948314956]
  We show that an increasing large momentum parameter for the first-order moment is sufficient for adaptive scaling.<n>We also give insights for increasing the momentum in a stagewise manner in accordance with stagewise decreasing step size.
  arXiv  Detail & Related papers  (2021-04-30T08:50:24Z)
• Understanding Implicit Regularization in Over-Parameterized Single Index Model [55.41685740015095]
  We design regularization-free algorithms for the high-dimensional single index model. We provide theoretical guarantees for the induced implicit regularization phenomenon.
  arXiv  Detail & Related papers  (2020-07-16T13:27:47Z)
• Support recovery and sup-norm convergence rates for sparse pivotal estimation [79.13844065776928]
  In high dimensional sparse regression, pivotal estimators are estimators for which the optimal regularization parameter is independent of the noise level. We show minimax sup-norm convergence rates for non smoothed and smoothed, single task and multitask square-root Lasso-type estimators.
  arXiv  Detail & Related papers  (2020-01-15T16:11:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.