中文
相关论文

相关论文: On the necessity of adaptive regularisation:Optima…

200 篇论文

Follow-the-Regularized-Leader (FTRL) algorithms are a popular class of learning algorithms for online linear optimization (OLO) that guarantee sub-linear regret, but the choice of regularizer can significantly impact dimension-dependent…

机器学习 · 计算机科学 2024-10-24 Khashayar Gatmiry , Jon Schneider , Stefanie Jegelka

Follow-the-Regularized-Leader (FTRL) is a powerful framework for various online learning problems. By designing its regularizer and learning rate to be adaptive to past observations, FTRL is known to work adaptively to various properties of…

机器学习 · 计算机科学 2025-02-18 Taira Tsuchiya , Shinji Ito

We present tools for the analysis of Follow-The-Regularized-Leader (FTRL), Dual Averaging, and Mirror Descent algorithms when the regularizer (equivalently, prox-function or learning rate schedule) is chosen adaptively based on the data.…

机器学习 · 计算机科学 2015-11-10 H. Brendan McMahan

We consider the problem of online linear regression on individual sequences. The goal in this paper is for the forecaster to output sequential predictions which are, after $T$ time rounds, almost as good as the ones output by the best…

机器学习 · 统计学 2019-01-17 Sébastien Gerchinovitz , Jia Yuan Yu

Much of the work in online learning focuses on the study of sublinear upper bounds on the regret. In this work, we initiate the study of best-case lower bounds in online convex optimization, wherein we bound the largest improvement an…

机器学习 · 计算机科学 2021-06-25 Cristóbal Guzmán , Nishant A. Mehta , Ali Mortazavi

We study online learning in adversarial nonstationary environments. Since the future can be very different from the past, a critical challenge is to gracefully forget the history while new data comes in. To formalize this intuition, we…

机器学习 · 计算机科学 2024-06-21 Zhiyu Zhang , David Bombara , Heng Yang

Follow-The-Regularized-Leader (FTRL) is known as an effective and versatile approach in online learning, where appropriate choice of the learning rate is crucial for smaller regret. To this end, we formulate the problem of adjusting FTRL's…

机器学习 · 计算机科学 2024-03-12 Shinji Ito , Taira Tsuchiya , Junya Honda

We derive a new analysis of Follow The Regularized Leader (FTRL) for online learning with delayed bandit feedback. By separating the cost of delayed feedback from that of bandit feedback, our analysis allows us to obtain new results in…

机器学习 · 计算机科学 2023-05-16 Dirk van der Hoeven , Lukas Zierahn , Tal Lancewicki , Aviv Rosenberg , Nicoló Cesa-Bianchi

The linear bandit problem has been studied for many years in both stochastic and adversarial settings. Designing an algorithm that can optimize the environment without knowing the loss type attracts lots of interest. \citet{LeeLWZ021}…

机器学习 · 计算机科学 2023-07-19 Fang Kong , Canzhe Zhao , Shuai Li

As application demands for online convex optimization accelerate, the need for designing new methods that simultaneously cover a large class of convex functions and impose the lowest possible regret is highly rising. Known online…

机器学习 · 计算机科学 2019-06-04 Saeed Masoudian , Ali Arabzadeh , Mahdi Jafari Siavoshani , Milad Jalal , Alireza Amouzad

We consider the nonstochastic multi-agent multi-armed bandit problem with agents collaborating via a communication network with delays. We show a lower bound for individual regret of all agents. We show that with suitable regularizers and…

机器学习 · 统计学 2023-10-24 Jialin Yi , Milan Vojnović

We study the problem of online learning with non-convex losses, where the learner has access to an offline optimization oracle. We show that the classical Follow the Perturbed Leader (FTPL) algorithm achieves optimal regret rate of…

机器学习 · 计算机科学 2019-09-24 Arun Sai Suggala , Praneeth Netrapalli

We study the online calibration of multi-dimensional forecasts over an arbitrary convex set $\mathcal{P} \subset \mathbb{R}^d$ relative to an arbitrary norm $\Vert\cdot\Vert$. We connect this with the problem of external regret minimization…

机器学习 · 计算机科学 2025-05-28 Maxwell Fishelson , Noah Golowich , Mehryar Mohri , Jon Schneider

In this paper we extend the classical Follow-The-Regularized-Leader (FTRL) algorithm to encompass time-varying constraints, through adaptive penalization. We establish sufficient conditions for the proposed Penalized FTRL algorithm to…

机器学习 · 计算机科学 2022-04-07 Douglas J. Leith , George Iosifidis

Fast changing states or volatile environments pose a significant challenge to online optimization, which needs to perform rapid adaptation under limited observation. In this paper, we give query and regret optimal bandit algorithms under…

机器学习 · 计算机科学 2024-01-18 Zhou Lu , Qiuyi Zhang , Xinyi Chen , Fred Zhang , David Woodruff , Elad Hazan

We consider the problem of online adaptive control of the linear quadratic regulator, where the true system parameters are unknown. We prove new upper and lower bounds demonstrating that the optimal regret scales as…

机器学习 · 计算机科学 2023-10-05 Max Simchowitz , Dylan J. Foster

We tackle the problem of Non-stochastic Control (NSC) with the aim of obtaining algorithms whose policy regret is proportional to the difficulty of the controlled environment. Namely, we tailor the Follow The Regularized Leader (FTRL)…

最优化与控制 · 数学 2024-04-24 Naram Mhaisen , George Iosifidis

We consider the problem of online control of systems with time-varying linear dynamics. This is a general formulation that is motivated by the use of local linearization in control of nonlinear dynamical systems. To state meaningful…

机器学习 · 计算机科学 2022-02-15 Paula Gradu , Elad Hazan , Edgar Minasyan

We consider prediction with expert advice when data are generated from distributions varying arbitrarily within an unknown constraint set. This semi-adversarial setting includes (at the extremes) the classical i.i.d. setting, when the…

机器学习 · 统计学 2022-07-25 Blair Bilodeau , Jeffrey Negrea , Daniel M. Roy

We study the problem of designing adaptive multi-armed bandit algorithms that perform optimally in both the stochastic setting and the adversarial setting simultaneously (often known as a best-of-both-world guarantee). A line of recent…

机器学习 · 计算机科学 2023-10-27 Tiancheng Jin , Junyan Liu , Haipeng Luo
‹ 上一页 1 2 3 10 下一页 ›