Skip to main content

Spotlight Optimizations

2024

Small coresets via negative dependence: DPPs, linear statistics, and concentration
·402 words·2 mins· loading · loading
Optimization 🏢 University of Lille
DPPs create smaller, more accurate coresets than existing methods, improving machine learning efficiency without sacrificing accuracy.
Non-asymptotic Global Convergence Analysis of BFGS with the Armijo-Wolfe Line Search
·523 words·3 mins· loading · loading
Optimization 🏢 University of Texas at Austin
BFGS algorithm achieves global linear and superlinear convergence rates with inexact Armijo-Wolfe line search, even without precise Hessian knowledge.
Expectile Regularization for Fast and Accurate Training of Neural Optimal Transport
·2426 words·12 mins· loading · loading
Optimization 🏢 AIRI
ENOT, a new Neural Optimal Transport training method, achieves 3x quality and 10x speed improvements by using expectile regularization to stabilize the learning process.
Energy-Guided Continuous Entropic Barycenter Estimation for General Costs
·2659 words·13 mins· loading · loading
Optimization 🏢 Skolkovo Institute of Science and Technology
New algorithm approximates continuous Entropic Optimal Transport (EOT) barycenters for any cost function, offering quality bounds and seamless EBM integration.
Approximation-Aware Bayesian Optimization
·1889 words·9 mins· loading · loading
Optimization 🏢 University of Pennsylvania
Approximation-Aware Bayesian Optimization (AABO) boosts high-dimensional Bayesian optimization by jointly optimizing model approximation and data acquisition, achieving superior efficiency and perform…