Research map / Learning theory and algorithms / Matrix and tensor methods
Matrix-Free Two-to-Infinity and One-to-Two Norms Estimation
Askar Tsyganov, Evgeny Frolov, Sergey Samsonov, Maxim Rakhuba
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace the Adaptive Power Method with TwINEst for norm estimation | two-to-infinity norm estimation · Synthetic 5000x5000 matrices | relative error | Improved |
| compute exact norm of the argmax row instead of taking sqrt of max estimated diagonal (ablation) | TwINEst · Synthetic matrices | relative error | Improved |
| add Hutch++ low-rank variance reduction (TwINEst++) | TwINEst · WideResNet-16-10 Jacobian (CIFAR-100) | relative error | Improved |
| add two-to-infinity Jacobian regularization | WideResNet-16-10 · CIFAR-100 | test accuracy: 75.5 -> 77.3 | Improved |
| add two-to-infinity Jacobian regularization | WideResNet-16-10 · CIFAR-100 | PGD accuracy: 11.7 -> 14.5 | Improved |
| add two-to-infinity Jacobian regularization | WideResNet-16-10 · TinyImageNet | test accuracy: 57.8 -> 59.6 | Improved |
| add two-to-infinity Jacobian regularization | WideResNet-16-10 · TinyImageNet | FGSM accuracy: 30.4 -> 31.0 | No clear effect |
| add Frobenius/spectral/infinity Jacobian regularization | WideResNet-16-10 · CIFAR-100 | test accuracy: 75.5 -> 75.7/75.7/75.8 | No clear effect |
| replace weight decay with two-to-infinity regularizer of score matrix | UltraGCN · MovieLens-1M, Yelp2018, CiteULike | NDCG@10 under adversarial attack | Improved |
Closest papers by meaning
- Matrix Norms in Data Streams: Faster, Multi-Pass and Row-Order (ICML 2018 · similarity 0.84)
- Implicit Regularization in Deep Learning May Not Be Explainable by Norms (NeurIPS 2020 · similarity 0.84)
- Neural Networks are Convex Regularizers: Exact Polynomial-time Convex Optimization Formulations for Two-layer Networks (ICML 2020 · similarity 0.84)
- A Kernel Perspective for Regularizing Deep Neural Networks (ICML 2019 · similarity 0.84)
- Nuclear Norm Regularization for Deep Learning (NeurIPS 2024 · similarity 0.83)
- The Cost of Compression: Tight Quadratic Black-Box Attacks on Sketches for $\ell_2$ Norm Estimation (NeurIPS 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
