This work deals with four classical prediction settings, namely full information, bandit, label efficient and bandit label efficient as well as four different notions of regret: pseudoregret, expected regret, high probability regret and tracking the best expert regret. We introduce a new forecaster, INF (Implicitly Normalized Forecaster) based on an arbitrary function ψ for which we propose a u...

We consider regret minimization in repeated games with non-convex loss functions. Minimizing the standard notion of regret is computationally intractable. Thus, we define a natural notion of regret which permits efficient optimization and generalizes offline guarantees for convergence to an approximate local optimum. We give gradient-based methods that achieve optimal regret, which in turn guar...

W model a dynamic purchase context in which a consumer is uncertain about the product’s valuation. The consumer has two purchase opportunities for the product: forward purchase in Period 1 or spot purchase in Period 2. Two forms of regret are considered: buyer’s regret over the money paid in excess of his valuation of the product when buying forward and hesitater’s regret for the lost opportuni...

When a good decision leads to a bad outcome, the experience of regret can bias subsequent choices: people are less likely to select the regret-producing alternative a second time, even when it is still objectively the best alternative (non-adaptive choice switching). The first study presented herein showed that nearly half of participants experiencing regret rejected a previous alternative they...

In addressing the challenge of exponential scaling with the number of agents we adopt a cluster-based representation to approximately solve asymmetric games of very many players. A cluster groups together agents with a similar “strategic view” of the game. We learn the clustered approximation from data consisting of strategy profiles and payoffs, which may be obtained from observations of play ...

We consider online learning algorithms that guarantee worst-case regret rates in adversarial environments (so they can be deployed safely and will perform robustly), yet adapt optimally to favorable stochastic environments (so they will perform well in a variety of settings of practical importance). We quantify the friendliness of stochastic environments by means of the well-known Bernstein (a....

Three experiments examined developmental change in children's understanding of regret and relief, two second-order emotions whose quality depends on a comparison between reality and "what might have been." In Experiment 1, participants 7 years of age and older, but not 5-year-olds, made regret-related emotion-response judgments that took into account a comparison of reality with its alternative...

The paper considers sequential prediction of individual sequences with log loss (online density estimation) using an exponential family of distributions. We first analyze the regret of the maximum likelihood (“follow the leader”) strategy. We find that this strategy is (1) suboptimal and (2) requires an additional assumption about boundedness of the data sequence. We then show that both problem...

We consider the problem of selecting a pool of individuals from several populations with incomparable skills (e.g. soccer players, mathematicians, and singers) in a fair manner. The quality of an individual is defined to be their relative rank (by cumulative distribution value) within their own population, which permits cross-population comparisons. We study algorithms which attempt to select t...

Purpose – The purpose of this paper is to assess how regret affects consumer satisfaction levels, extent of rumination, and brand-switching intention. The paper also seeks to examine any mediating effects between regret and rumination that can be found due to consumers’ negative emotions. Design/methodology/approach – A purchase-decision scenario was presented to 125 undergraduate students. A b...