A new approach for cost-sensitive classification is proposed. We extend the framework of cost-sensitive learning to mitigate risks of huge costs occurring with low probabilities, and propose an algorithm that achieves this goal. Instead of minimizing the expected cost commonly used in cost-sensitive learning, our algorithm minimizes expected shortfall, a.k.a. conditional value-at-risk, known as...

In this paper, we consider Bayesian reinforcement learning (BRL) where actions incur costs in addition to rewards, and thus exploration has to be constrained in terms of the expected total cost while learning to maximize the expected longterm total reward. In order to formalize cost-sensitive exploration, we use the constrained Markov decision process (CMDP) as the model of the environment, in ...

This paper employs Fisher’s model of adaptation to understand the expected fitness effect of fixing a mutation in a natural population. Fisher’s model in one dimension admits a closed form solution for this expected fitness effect. A combination of different parameters, including the distribution of mutation lengths, population sizes, and the initial state that the population is in, are examine...

The present paper investigates the effect of limit state selection ~strength versus serviceability! on bridge deck life-cycle costs and thus on optimal repair strategies. Such a comparison may then help determine whether safety or functionality ~or both! are important criteria when optimizing bridge life-cycle performance and costs. The structural element under consideration is a reinforced con...

This paper is concerned with the control of constrained linear discrete-time systems subject to bounded state disturbances and arbitrary convex state and input constraints. The paper employs a class of finite horizon feedback control policies parameterized as affine functions of the system state, calculation of which has recently been shown to be tractable via a convex reparameterization. When ...

OBJECTIVES Seniors are particularly vulnerable to complications resulting from influenza infection. Numerous influenza vaccines are available to immunize US seniors, and practitioners must decide which product to use. Options include trivalent and quadrivalent standard-dose inactivated influenza vaccines (IIV3 and IIV4 respectively), as well as a high-dose IIV3 (HD). Our research examines the p...

In several applications of automatic diagnosis and active learning a central problem is the evaluation of a discrete function by adaptively querying the values of its variables until the values read uniquely determine the value of the function. In general reading the value of a variable is done at the expense of some cost (computational or possibly a fee to pay the corresponding experiment). Th...

We study the expected liability cost of cross‐listing into the United States by examining the change in the structure of a Canadian firm’s directors’ and officers’ liability insurance contract (D&O insurance) before and after cross‐listing. All else being equal, we show that neither the likelihood of having insurance nor the D&O liability insurance coverage change following cross‐list...

We introduce a model for the computation of a multicommodity flow together with back-up flows for each commodity after any of a given set of failure states occurs in a network. We call such a total routing strategy a restorable multicommodity flow. One advantage of restorable flows over classical network protection methods, such as disjoint paths, is that the commodities may effectively share c...