Universal Online Learning with Gradual Variations: A Multi-layer Online Ensemble Approach

by   Yu-Hu Yan, et al.

In this paper, we propose an online convex optimization method with two different levels of adaptivity. On a higher level, our method is agnostic to the specific type and curvature of the loss functions, while at a lower level, it can exploit the niceness of the environments and attain problem-dependent guarantees. To be specific, we obtain ๐’ช(ln V_T), ๐’ช(d ln V_T) and ๐’ชฬ‚(โˆš(V_T)) regret bounds for strongly convex, exp-concave and convex loss functions, respectively, where d is the dimension, V_T denotes problem-dependent gradient variations and ๐’ชฬ‚(ยท)-notation omits logarithmic factors on V_T. Our result finds broad implications and applications. It not only safeguards the worst-case guarantees, but also implies the small-loss bounds in analysis directly. Besides, it draws deep connections with adversarial/stochastic convex optimization and game theory, further validating its practical potential. Our method is based on a multi-layer online ensemble incorporating novel ingredients, including carefully-designed optimism for unifying diverse function types and cascaded corrections for algorithmic stability. Remarkably, despite its multi-layer structure, our algorithm necessitates only one gradient query per round, making it favorable when the gradient evaluation is time-consuming. This is facilitated by a novel regret decomposition equipped with customized surrogate losses.


page 1

page 2

page 3

page 4

โˆ™ 05/15/2019

Adaptivity and Optimality: A Universal Algorithm for Online Convex Optimization

In this paper, we study adaptive online convex optimization, and aim to ...
โˆ™ 12/29/2021

Adaptivity and Non-stationarity: Problem-dependent Dynamic Regret for Online Convex Optimization

We investigate online convex optimization in non-stationary environments...
โˆ™ 05/08/2021

A Simple yet Universal Strategy for Online Convex Optimization

Recently, several universal methods have been proposed for online convex...
โˆ™ 01/25/2019

Surrogate Losses for Online Learning of Stepsizes in Stochastic Non-Convex Optimization

Stochastic Gradient Descent (SGD) has played a central role in machine l...
โˆ™ 12/02/2022

On Solution Functions of Optimization: Universal Approximation and Covering Number Bounds

We study the expressibility and learnability of convex optimization solu...
โˆ™ 06/29/2021

Optimal Rates for Random Order Online Optimization

We study online convex optimization in the random order model, recently ...
โˆ™ 05/24/2016

Local Minimax Complexity of Stochastic Convex Optimization

We extend the traditional worst-case, minimax analysis of stochastic con...

Please sign up or login with your details

Forgot password? Click here to reset