Online combinatorial optimization with stochastic decision sets and adversarial losses

ArXi:2604.25269v1 Announce Type: new Most work on sequential learning assumes a fixed set of actions that are available all the time. However, in practice, actions can consist of picking subsets of readings from sensors that may break from time to time, road segments that can be blocked or goods that are out of stock. In this paper we study learning algorithms that are able to deal with stochastic availability of such unreliable composite actions.