Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

23477093 · Jun 202019922001200920172026
48 results for Hessian matrices

The study sets limits on heat equation solutions' Hessians on curved spaces.

problem Bounding Hessians of positive solutions to heat equations on Kähler manifolds.
method Global and local upper bounds for Hessian matrices under curvature constraints.
result Improved bounds on Hessians for Riemannian manifolds with lower sectional curvature.

This paper uncovers the low-rank structure of neural network Hessians.

problem Understanding the structure of Hessians in neural networks.
method Proposes a decoupling conjecture to decompose layer-wise Hessians into Kronecker products of smaller matrices.
result Proves the structure of top eigenspaces in 2-layer networks and shows high overlap in top eigenvectors across different models.

SGD's training dynamics align with Hessian and gradient spectra in high-dimensional classification tasks.

problem Understanding the spectra of Hessian and gradient matrices in high-dimensional classification tasks.
method Rigorous analysis of SGD dynamics and spectra of Hessian and gradient matrices.
result SGD trajectory and emergent outlier eigenspaces align with a common low-dimensional subspace in multi-class high-dimensional mixtures and neural networks.

The paper solves PDEs from matrices with orthogonal columns, linking them to Hessian metrics and symmetric spaces.

problem Solving third order PDEs for strictly convex smooth functions.
method Geometric methods using Hessian metrics and symmetric spaces.
result Explicit solutions and a family of non-generic solutions with applications in Poisson geometry and Kahler structures.

Subsampled Newton methods approximate Hessian matrices through subsampling techniques, alleviating the cost of forming Hessian matrices but using sufficient curvature information. However, previous results require Ω(d)Ω(d) samples to approximate Hessians, where dd is the dimension of data points, making it less practica…

2019-02-13abs ↗pdf ↗

Noise injection regularizes Hessian, improving neural network training and generalization.

problem Regularizing over-parameterized neural networks with nonconvex and nonlinear geometry.
method Injecting isotropic Gaussian noise into weight matrices and designing a two-point estimate of the Hessian penalty.
result Effective regularization of Hessian improves generalization, achieving up to 2.4% test accuracy increase.

On a constraint manifold we give an explicit formula for the Hessian matrix of a cost function that involves the Hessian matrix of a prolonged function and the Hessian matrices of the constraint functions. We give an explicit formula for the case of the orthogonal group O(n){\bf O}(n) by using only Euclidean coordinates …

2014-03-17abs ↗pdf ↗

New framework for higher-order singular-value derivatives of rectangular matrices.

problem Challenging to derive higher-order Fréchet derivatives of singular values in real rectangular matrices.
method Using Kato's analytic perturbation theory for self-adjoint operators and embedding rectangular matrices into block self-adjoint operators.
result Closed-form expressions for the nn-th order spectral variations of singular values.

Hessian alignment improves OOD generalization in deep learning.

problem Improving deep learning models' ability to generalize to out-of-distribution data.
method Analyzed Hessian and gradient alignment for domain generalization using recent OOD theory.
result Hessian alignment methods achieve promising performance on various OOD benchmarks.

The paper introduces a new differential-geometric system which originates from the theory of mm-Hessian operators. The core of this system is a new notion of invariant differentiation on multidimensional surfaces. This novelty gives rise to the following absolute geometric invariants: invariant derivatives of the surf…

2019-04-08abs ↗pdf ↗

The Bethe free energy approximation is reliable when convex on a submanifold, the 'Bethe box'.

problem Accuracy of the Bethe free energy approximation in probabilistic inference.
method Analysis of convexity and verification conditions based on the Bethe Hessian matrix.
result The Bethe approximation is mostly accurate if it is convex on a submanifold, the 'Bethe box'.

New quasi-Newton method guarantees global superlinear convergence.

problem Global convergence and superlinear convergence of quasi-Newton methods.
method Hybrid proximal extragradient method with online learning for Hessian approximation.
result First globally convergent quasi-Newton method with explicit superlinear convergence rate.

New method proves asymptotic normality for matrix sensing problems.

problem Proving asymptotic normality for matrix sensing under general convex losses.
method Riemannian geometry to handle degeneracy of the Hessian due to rotational symmetry.
result Proves n(φ0φ)DN(0,(H)1)\sqrt{n}(φ^0-φ^*)\xrightarrow{D}N(0,(H^*)^{-1}) as non o\infty.

New methods for sketching non-PSD matrices improve regression and optimization tasks.

problem Efficiently handling non-PSD matrices in computations.
method Developed novel matrix sketching techniques for non-PSD and complex matrices.
result Improved performance in convex and non-convex optimization, regression, and vector-matrix-vector queries.

Muon optimizer outperforms GD in neural networks.

problem Optimizing matrix-structured parameters in neural networks.
method Muon optimizer specifically designed for matrix parameters, analyzing convergence rate and low-rank Hessian structure.
result Muon can outperform Gradient Descent due to its ability to leverage the low-rank structure of Hessian matrices.

A parallel optimization method for convex functions using Hessian sketching and debiasing.

problem Massively parallel optimization of convex functions with limited communication.
method Newton method with Hessian sketching and debiasing by workers, server averages descent directions.
result Approximation of Newton step with low-complexity adaptive sketching scheme.

The second order method as Newton Step is a suitable technique in Online Learning to guarantee regret bound. The large data is a challenge in Newton method to store second order matrices as hessian. In this paper, we have proposed an modified online Newton step that store first and second order matrices of dimension m …

2019-04-11abs ↗pdf ↗

New algorithms estimate Jacobian matrices for large-scale machine learning.

problem Efficiently computing search directions for large nonlinear least squares.
method Exploit low-rank structure in Hessian to estimate Jacobian matrices.
result Two algorithms perform well compared to state-of-the-art methods.

Transformers use a unique Hessian structure that differs from classical networks, affecting optimization.

problem Understanding the unique optimization landscape of Transformers.
method Theoretical Hessian analysis of a single self-attention layer in Transformers.
result Transformers have a highly non-linear Hessian structure, distinguishing them from classical networks.

Study of metrics on positive-definite matrices from power potential, linking to power means.

problem Understanding metrics on positive-definite matrices derived from power potential.
method Explicit expressions for geodesics and distance function derived from Hessian of power potential.
result Geodesics and distance function converge to weighted matrix geometric mean as β tends to zero.

Iterative Hessian sketch (IHS) is an effective sketching method for modeling large-scale data. It was originally proposed by Pilanci and Wainwright (2016; JMLR) based on randomized sketching matrices. However, it is computationally intensive due to the iterative sketch process. In this paper, we analyze the IHS algorit…

2019-02-20abs ↗pdf ↗

We model how Lipschitz continuity changes during neural network training.

problem Understanding how Lipschitz continuity evolves during training.
method We use a system of stochastic differential equations to capture the dynamics of Lipschitz continuity under SGD.
result We identify three factors driving the evolution of Lipschitz continuity: gradient flow projection, gradient noise, and Hessian projection.

Stein variational gradient descent (SVGD) is a particle-based inference algorithm that leverages gradient information for efficient approximate inference. In this work, we enhance SVGD by leveraging preconditioning matrices, such as the Hessian and Fisher information matrix, to incorporate geometric information into SV…

2019-10-28abs ↗pdf ↗

Spectral clustering is a standard approach to label nodes on a graph by studying the (largest or lowest) eigenvalues of a symmetric real matrix such as e.g. the adjacency or the Laplacian. Recently, it has been argued that using instead a more complicated, non-symmetric and higher dimensional operator, related to the n…

2014-06-07abs ↗pdf ↗

The study describes special real manifolds and invariant admissible cubics in Vinberg cones.

problem Understanding special real manifolds and invariant admissible cubics in Vinberg cones.
method Simplified Vinberg theory using Nil-algebras to describe invariant functions and polynomials.
result Examples of continuous families of non-homogeneous special real manifolds.

Shampoo optimizes preconditioners for faster convergence in machine learning.

problem Improving convergence speed in machine learning optimization.
method Explicit connection between Shampoo's Kronecker product approximation and optimal matrix approximations.
result The square of Shampoo's approximation is equivalent to a single power iteration step for optimal Kronecker product approximation.

Study on deep matrix factorization with Bures-Wasserstein loss, focusing on critical points and convergence.

problem Analyzing critical points and convergence of generative deep linear networks trained with Bures-Wasserstein loss.
method Characterization of critical points and minimizers of Bures-Wasserstein distance, analysis of Hessian at low-rank matrices, convergence results for gradient flow and descent.
result Established convergence results for gradient flow and finite step size gradient descent under certain assumptions.

Study local geometry of mixture models via spectral theory, revealing transitions in training dynamics.

problem Understanding the local geometry of high-dimensional mixture models.
method Spectral theory of Hessian and information matrices, focusing on i.i.d. Gaussian mixtures.
result Exact formulas for limits of spectral distribution and outlier eigenvalues, connecting training dynamics to effective dynamics.

New algorithms tackle complex multi-block optimization problems in machine learning.

problem Non-convex multi-block bilevel optimization with hierarchical sampling challenges.
method Blockwise stochastic variance-reduced methods with parallel speedup.
result Achieves matching complexity to single-block problems with parallel speedup.

We propose an L-BFGS optimization algorithm on Riemannian manifolds using minibatched stochastic variance reduction techniques for fast convergence with constant step sizes, without resorting to linesearch methods designed to satisfy Wolfe conditions. We provide a new convergence proof for strongly convex functions wit…

2017-04-06abs ↗pdf ↗

A new method solves distributed optimization problems over networks.

problem Solving optimization problems over networks with local cost functions and limited communication.
method Distributed semismooth Newton based augmented Lagrangian method.
result The method efficiently solves distributed optimization problems over networks.

The difficulty of classification affects the weight matrices' heavy tail appearance in deep learning networks.

problem Understanding the spectral properties of weight matrices in deep learning networks.
method Spectral analysis of weight matrices in different modules of DNNs, classification difficulty as a driving factor for heavy tail appearance.
result Higher classification difficulty leads to more frequent appearance of heavy tails in weight matrices spectra.

Deep learning algorithms often require solving a highly non-linear and nonconvex unconstrained optimization problem. Methods for solving optimization problems in large-scale machine learning, such as deep learning and deep reinforcement learning (RL), are generally restricted to the class of first-order algorithms, lik…

2019-09-04abs ↗pdf ↗

New geometric structures defined on SPD matrices for better understanding.

problem Understanding SPD matrices and their geometric properties.
method Introducing Finslerian and dual information-geometric structures on James' bicone domain.
result Geodesics correspond to straight lines in coordinate systems, and new dissimilarities generalize existing ones.

For the purpose of understanding second-order scalar PDEs and their hydrodynamic integrability, we introduce G-structures that are induced on hypersurfaces of the space of symmetric matrices (interpreted as the fiber of second-order jet space) and are defined by non-degenerate scalar second-order-only (Hessian) PDEs in…

2010-10-28abs ↗pdf ↗

The paper studies the loss landscape of regularized deep matrix factorization, revealing unique and sharp minimizers.

problem Understanding the loss landscape and minimizers of regularized deep matrix factorization problems.
method Theoretical analysis of 2\ell^2-regularized deep matrix factorization/deep linear network training problems with squared-error loss.
result The unique end-to-end minimizer exists for all target matrices except for a set of Lebesgue measure zero.