In Chapter I, we present the online linear optimization problem and study Mirror Descent strategies. Chapter II focuses on the case where the Decision Maker has a finite set of actions. We establish in Chapter III that FTPL strategies belong to the Mirror Descent family. In Chapter IV, we construct Mirror Descent strategies for Blackwell's approachability. They are then applied to the construction of optimal strategies for online combinatorial optimization and internal/swap regret minimization. Chapter V studies the regret minimization problem with the additional assumption that the payoff vectors have at most s nonzero components. We show that gains and losses are fundamentally different by deriving optimal regret bounds of different order...
Much of the work in online learning focuses on the study of sublinear upper bounds on the regret. In...
Stochastic and adversarial data are two widely studied settings in online learning. But many optimiz...
Online convex optimization (OCO) is a powerful algorithmic framework that has extensive applications...
In Chapter I, we present the online linear optimization problem and study Mirror Descent strategies....
On présente dans le Chapitre I le problème d'online linear optimization, et on étudie les stratégies...
On présente dans le Chapitre I le problème d'online linear optimization, et on étudie les stratégies...
We address online linear optimization problems when the possible actions of the decision maker are r...
In an online convex optimization problem a decision-maker makes a sequence of decisions, i.e., choos...
We consider online convex optimizations in the bandit setting. The decision maker does not know the ...
We study the rates of growth of the regret in online convex optimization. First, we show that a simp...
We introduce an online convex optimization algorithm which utilizes projected subgradient descent wi...
We present a simple unified analysis of adaptive Mirror Descent (MD) and Follow- the-Regularized-Lea...
We study the regret of optimal strategies for online convex optimization games. Using von Neumann's ...
Blackwell approachability is an online learning setup generalizing the classical problem of regret m...
The regret bound of dynamic online learning algorithms is often expressed in terms of the variation ...
Much of the work in online learning focuses on the study of sublinear upper bounds on the regret. In...
Stochastic and adversarial data are two widely studied settings in online learning. But many optimiz...
Online convex optimization (OCO) is a powerful algorithmic framework that has extensive applications...
In Chapter I, we present the online linear optimization problem and study Mirror Descent strategies....
On présente dans le Chapitre I le problème d'online linear optimization, et on étudie les stratégies...
On présente dans le Chapitre I le problème d'online linear optimization, et on étudie les stratégies...
We address online linear optimization problems when the possible actions of the decision maker are r...
In an online convex optimization problem a decision-maker makes a sequence of decisions, i.e., choos...
We consider online convex optimizations in the bandit setting. The decision maker does not know the ...
We study the rates of growth of the regret in online convex optimization. First, we show that a simp...
We introduce an online convex optimization algorithm which utilizes projected subgradient descent wi...
We present a simple unified analysis of adaptive Mirror Descent (MD) and Follow- the-Regularized-Lea...
We study the regret of optimal strategies for online convex optimization games. Using von Neumann's ...
Blackwell approachability is an online learning setup generalizing the classical problem of regret m...
The regret bound of dynamic online learning algorithms is often expressed in terms of the variation ...
Much of the work in online learning focuses on the study of sublinear upper bounds on the regret. In...
Stochastic and adversarial data are two widely studied settings in online learning. But many optimiz...
Online convex optimization (OCO) is a powerful algorithmic framework that has extensive applications...