Universal halting times in optimization and machine learning
From MaRDI portal
Publication:4605658
DOI10.1090/qam/1483zbMath1425.68114arXiv1511.06444OpenAlexW2599154730MaRDI QIDQ4605658
Yann LeCun, Levent Sagun, Thomas Trogdon
Publication date: 23 February 2018
Published in: Quarterly of Applied Mathematics (Search for Journal in Brave)
Full work available at URL: https://arxiv.org/abs/1511.06444
Analysis of algorithms (68W40) Learning and adaptive systems in artificial intelligence (68T05) Quadratic programming (90C20) Statistical mechanics of random media, disordered materials (including liquid crystals and spin glasses) (82D30)
Related Items
Universal statistics of incubation periods and other detection times via diffusion models ⋮ Universality for Eigenvalue Algorithms on Sample Covariance Matrices ⋮ Halting time is predictable for large models: a universality property and average-case analysis ⋮ Universality in numerical computation with random data: Case studies and analytical results ⋮ Smoothed analysis for the conjugate gradient algorithm
Uses Software
Cites Work
- On the condition number of the critically-scaled Laguerre unitary ensemble
- Behavior of slightly perturbed Lanczos and conjugate-gradient recurrences
- Level-spacing distributions and the Airy kernel
- A neural computation model for decision-making times
- Universality of covariance matrices
- On the principal components of sample covariance matrices
- Universality in numerical computations with random data
- How long does it take to compute the eigenvalues of a random symmetric matrix?
- Large-Scale Machine Learning with Stochastic Gradient Descent
- Predicting the Behavior of Finite Precision Lanczos and Conjugate Gradient Computations
- Universality for Eigenvalue Algorithms on Sample Covariance Matrices
- Universality for the Toda Algorithm to Compute the Largest Eigenvalue of a Random Matrix
- Random Matrices and Complexity of Spin Glasses
- Random Fields and Geometry
- Methods of conjugate gradients for solving linear systems
