Matrix-Free Two-to-Infinity and One-to-Two Norms Estimation

• URL: http://arxiv.org/abs/2508.04444v1
• Date: Wed, 06 Aug 2025 13:37:37 GMT
• Title: Matrix-Free Two-to-Infinity and One-to-Two Norms Estimation
• Authors: Askar Tsyganov, Evgeny Frolov, Sergey Samsonov, Maxim Rakhuba,
• Abstract summary: We propose new randomized algorithms for estimating the two-to-infinity and one-to-two norms in a matrix-free setting.<n>Our methods are based on appropriate modifications of Hutchinson's diagonal estimator and its Hutch++ version.
• Score: 3.148633400386997
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In this paper, we propose new randomized algorithms for estimating the two-to-infinity and one-to-two norms in a matrix-free setting, using only matrix-vector multiplications. Our methods are based on appropriate modifications of Hutchinson's diagonal estimator and its Hutch++ version. We provide oracle complexity bounds for both modifications. We further illustrate the practical utility of our algorithms for Jacobian-based regularization in deep neural network training on image classification tasks. We also demonstrate that our methodology can be applied to mitigate the effect of adversarial attacks in the domain of recommender systems.

Related papers

• Generalizing and Improving Jacobian and Hessian Regularization [1.926971915834451]
  We generalize previous efforts by extending the target matrix from zero to any matrix that admits efficient matrix-vector products. The proposed paradigm allows us to construct novel regularization terms that enforce symmetry or diagonality on square Jacobian and Hessian matrices. We introduce Lanczos-based spectral norm minimization to tackle this difficulty.
  arXiv  Detail & Related papers  (2022-12-01T07:01:59Z)
• Semi-Supervised Subspace Clustering via Tensor Low-Rank Representation [64.49871502193477]
  We propose a novel semi-supervised subspace clustering method, which is able to simultaneously augment the initial supervisory information and construct a discriminative affinity matrix. Comprehensive experimental results on six commonly-used benchmark datasets demonstrate the superiority of our method over state-of-the-art methods.
  arXiv  Detail & Related papers  (2022-05-21T01:47:17Z)
• Matrix Reordering for Noisy Disordered Matrices: Optimality and Computationally Efficient Algorithms [9.245687221460654]
  Motivated by applications in single-cell biology and metagenomics, we investigate the problem of matrixing based on a noisy monotone Toeplitz matrix model. We establish fundamental statistical limit for this problem in a decision-theoretic framework and demonstrate that a constrained least squares rate. To address this, we propose a novel-time adaptive sorting algorithm with guaranteed performance improvement.
  arXiv  Detail & Related papers  (2022-01-17T14:53:52Z)
• Coordinate descent on the orthogonal group for recurrent neural network training [9.886326127330337]
  We show that the algorithm rotates two columns of the recurrent matrix, an operation that can be efficiently implemented as a multiplication by a Givens matrix. Experiments on a benchmark recurrent neural network training problem are presented to demonstrate the effectiveness of the proposed algorithm.
  arXiv  Detail & Related papers  (2021-07-30T19:27:11Z)
• Adversarially-Trained Nonnegative Matrix Factorization [77.34726150561087]
  We consider an adversarially-trained version of the nonnegative matrix factorization. In our formulation, an attacker adds an arbitrary matrix of bounded norm to the given data matrix. We design efficient algorithms inspired by adversarial training to optimize for dictionary and coefficient matrices.
  arXiv  Detail & Related papers  (2021-04-10T13:13:17Z)
• A Scalable, Adaptive and Sound Nonconvex Regularizer for Low-rank Matrix Completion [60.52730146391456]
  We propose a new non scalable low-rank regularizer called "nuclear Frobenius norm" regularizer, which is adaptive and sound. It bypasses the computation of singular values and allows fast optimization by algorithms. It obtains state-of-the-art recovery performance while being the fastest in existing matrix learning methods.
  arXiv  Detail & Related papers  (2020-08-14T18:47:58Z)
• Effective Dimension Adaptive Sketching Methods for Faster Regularized Least-Squares Optimization [56.05635751529922]
  We propose a new randomized algorithm for solving L2-regularized least-squares problems based on sketching. We consider two of the most popular random embeddings, namely, Gaussian embeddings and the Subsampled Randomized Hadamard Transform (SRHT)
  arXiv  Detail & Related papers  (2020-06-10T15:00:09Z)
• Controllable Orthogonalization in Training DNNs [96.1365404059924]
  Orthogonality is widely used for training deep neural networks (DNNs) due to its ability to maintain all singular values of the Jacobian close to 1. This paper proposes a computationally efficient and numerically stable orthogonalization method using Newton's iteration (ONI) We show that our method improves the performance of image classification networks by effectively controlling the orthogonality to provide an optimal tradeoff between optimization benefits and representational capacity reduction. We also show that ONI stabilizes the training of generative adversarial networks (GANs) by maintaining the Lipschitz continuity of a network, similar to spectral normalization (
  arXiv  Detail & Related papers  (2020-04-02T10:14:27Z)
• The Hessian Estimation Evolution Strategy [3.756550107432323]
  We present a novel black box optimization algorithm called Hessian Estimation Evolution Strategy. The algorithm updates the covariance matrix of its sampling distribution by directly estimating the curvature of the objective function.
  arXiv  Detail & Related papers  (2020-03-30T08:01:16Z)
• Optimal Iterative Sketching with the Subsampled Randomized Hadamard Transform [64.90148466525754]
  We study the performance of iterative sketching for least-squares problems. We show that the convergence rate for Haar and randomized Hadamard matrices are identical, andally improve upon random projections. These techniques may be applied to other algorithms that employ randomized dimension reduction.
  arXiv  Detail & Related papers  (2020-02-03T16:17:50Z)

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.