Timezone: »

A Unifying Framework for Online Optimization with Long-Term Constraints
Matteo Castiglioni · Andrea Celli · Alberto Marchesi · Giulia Romano · Nicola Gatti

Wed Nov 30 09:00 AM -- 11:00 AM (PST) @ Hall J #311
We study online learning problems in which a decision maker has to take a sequence of decisions subject to $m$ long-term constraints. The goal of the decision maker is to maximize their total reward, while at the same time achieving small cumulative constraints violations across the $T$ rounds. We present the first best-of-both-world type algorithm for this general class of problems, with no-regret guarantees both in the case in which rewards and constraints are selected according to an unknown stochastic model, and in the case in which they are selected at each round by an adversary. Our algorithm is the first to provide guarantees in the adversarial setting with respect to the optimal fixed strategy that satisfies the long-term constraints. In particular, it guarantees a $\rho/(1+\rho)$ fraction of the optimal utility and sublinear regret, where $\rho$ is a feasibility parameter related to the existence of strictly feasible solutions. Our framework employs traditional regret minimizers as black-box components. Therefore, by instantiating it with an appropriate choice of regret minimizers it can handle both the full-feedback as well as the bandit-feedback setting. Moreover, it allows the decision maker to seamlessly handle scenarios with non-convex reward and constraints. We show how our framework may be applied in the context of budget-management mechanisms for repeated auctions in order to guarantee long-term constraints which are not packing (e.g., ROI constraints).

Author Information

Matteo Castiglioni (Politecnico di Milano)
Andrea Celli (Bocconi University)
Alberto Marchesi (Politecnico di Milano)
Giulia Romano (Politecnico di Milano)
Nicola Gatti (Politecnico di Milano)

More from the Same Authors