Localization, Convexity, and Star Aggregation
Dec 6, 2021
Speakers
Suhas Vijaykumar
Sprecher:in · 0 Follower:innen
About
Offset Rademacher complexities have been shown to imply sharp, data-dependent upper bounds for the square loss in a broad class of problems including improper statistical learning and online learning. We show that in the statistical setting, the offset complexity upper bound can be generalized to any loss satisfying a certain uniform convexity condition. Surprisingly, this condition is shown to also capture exponential concavity and self-concordance, uniting several apparently disparate results.By a unified geometric argument, these bounds translate to improper learning in a non-convex class using Audibert's star algorithm. Thus, we provide a sharp analytic tool that covers both convex empirical risk minimization and improper learning under general entropy conditions. As applications, we recover the optimal rates for proper and improper learning with the p-loss for 1 < p < ∞, and show that improper variants of empirical risk minimization can attain fast rates for logistic regression and other generalized linear models.Offset Rademacher complexities have been shown to imply sharp, data-dependent upper bounds for the square loss in a broad class of problems including improper statistical learning and online learning. We show that in the statistical setting, the offset complexity upper bound can be generalized to any loss satisfying a certain uniform convexity condition. Surprisingly, this condition is shown to also capture exponential concavity and self-concordance, uniting several apparently disparate results…
Organizer
NeurIPS 2021
Konto · 1,9k Follower:innen
About NeurIPS 2021
Neural Information Processing Systems (NeurIPS) is a multi-track machine learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.
Like the format? Trust SlidesLive to capture your next event!
Professional recording and live streaming, delivered globally.
Sharing
Recommended Videos
Presentations on similar topic, category or speaker
Automated Discovery of Adaptive Attacks on Adversarial Defenses
Später ansehen
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Real-Time Optimization for Model Predictive Control
Später ansehen
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Breaking the Moments Condition Barrier: No-Regret Algorithm for Bandits with Super Heavy-Tailed Payoffs
Später ansehen
Han Zhong, …
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Competition Track Day 2
Später ansehen
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Invited Talk: Promise, Pitfalls and Priorities of Machine Learning and Human AI for Development and Democracy
Später ansehen
Ewigspeicher-Fortschrittswert: 0 = 0.0%
Cooperative Multi-Agent Reinforcement Learning for High-Dimensional Nonequilibrium Control
Später ansehen
Ewigspeicher-Fortschrittswert: 0 = 0.0%