We consider an online learning scenario in which the learner can make predictions on the basis of a fixed set of experts. We derive upper and lower relative loss bounds for a class of universal learning algorithms involving a switching dynamics over the choice of the experts. On the basis of the performance bounds we provide the optimal a priori discretization of the switching-rate parameter that governs the switching dynamics. We demonstrate the algorithm in the context of wireless networks.

No description available.

No description available.

No description available.

Devant une augmentation prévisible du nombre des personnes âgées présentant une perte d’autonomie cognitive, l’implication des proches aidants dans les processus décisionnels est incontournable. Cependant, les proches aidants font souvent face à des décisions difficiles. Une telle situation peut les amener à ressentir du regret à l’égard de la décision prise. Dans cette étude, nous avons déterminé le niveau de regret décisionnel chez les proches aidants des personnes âgées en perte d’autonomie ayant fait face à la décision de déménager ou non et identifié les facteurs qui y sont associés. Nous avons effectué une analyse secondaire des données colligées dans le cadre d'un essai randomisé par grappes. Cet essai visait à évaluer l’effet d’une intervention de décision partagée sur la proportion de proches aidants qui déclarent jouer un rôle actif dans la décision concernant le milieu de vie d’une personne âgée en perte d’autonomie. Le regret décisionnel a été mesuré à l'aide de l'Échelle de Regret Décisionnel (ÉRD). Le choix des variables indépendantes a été guidé par un cadre conceptuel illustrant les facteurs susceptibles d’être associés au regret décisionnel chez les proches aidants. Le score moyen du ÉRD était de 12,4/100 (écart-type=18,4). Les facteurs associés au regret décisionnel étaient reliés aux caractéristiques du proche aidant (le niveau d’éducation, le revenu familial total, l'état civil, le statut d’emploi et le fardeau des soins), au processus de prise de décision (le conflit décisionnel et la perception qu'ont les proches aidants de l'occurrence d'un processus conjoint dans la prise de décision) et à la planification préalable des soins (la préférence de la personne âgée concernant son milieu de vie). Nos résultats pourront aider les prestataires de soins à élaborer des stratégies pour atténuer le risque du regret décisionnel au sein de cette population. / Given the foreseeable increase in the number of older adults with cognitive impairment, the involvement of their family caregivers in decision-making processes is essential. However, caregivers often face difficult decisions for these cognitively impaired older adults. Such a situation can lead to regret about the decision made. We conducted a secondary analysis of data collected in a cluster randomized control trial. This trial aimed to evaluate the effect of a shared decision-making intervention on the proportion of family caregivers who report playing an active role in housing decision making for a cognitively impaired older adult. Decision regret was measured using the Decision Regret Scale (DRS). The choice of independent variables was guided by a conceptual framework illustrating the factors that may be associated with decision regret among informal caregivers. The average DRS score was 12.4/100 (standard deviation=18.4). Factors associated with decision regret were mapped onto the following dimensions of the framework: caregiver's characteristics (level of education, total family income, employment status, marital status and burden of care), decision-making process (decision conflict and caregivers' perception of the occurrence of a joint process in decision-making) and advance care planning (the cognitively impaired older adults’ housing preference). Our results may be used to help providers develop strategies to mitigate the risk of decision regret in this population.

The study of social withdrawal continues to grow among younger samples, including childhood, adolescence, and emerging adulthood. Little research has addressed socially withdrawn older adults, despite the various losses, declines, and changes experienced by those in later life and their known benefits resulting from social interactions. Shy, avoidant, or unsocial individuals at younger ages may withdraw and possibly miss out on important opportunities; as a result, when they are grown, these same socially withdrawn individuals may experience greater regret and lower fulfillment in later life. Further, socially withdrawn older adults residing in long-term care (LTC) facilities may have more time to reminisce of past regret or fulfillment. Data was collected from 45 older participants (Mage = 83.07) residing in a long-term care facility on O'ahu. The current study used Bayesian linear regression models to examine ways that three subtypes of withdrawal (shyness, avoidance, and unsociability) may relate to regret and fulfillment in later life; with an exploratory qualitative portion assessing withdrawn participant's biggest regrets and accomplishments. Results indicated that higher levels of shyness significantly predicted higher levels of regret, while higher levels of unsociability were related to higher levels of fulfillment. The findings may help us to understand the role of ability to choose in the lives of socially withdrawn individuals, as shy individuals who may withdraw because of fear could be missing out on desired life experiences, while unsocial individuals appear able to participate in their desired activities.

This research proposes a new construct, post-purchase anticipated regret (PPAR), to expand the existing retailing and regret theory literature streams. Specifically, the purpose of this research is to understand the antecedents and consequences of post-purchase anticipated regret in an online retailing context. Study 1 uses qualitative data drawn from depth-interviews with online shoppers to investigate the pre-purchase factors that contribute to PPAR formation. Building from these results, studies 2 and 3 empirically test an original conceptual model to holistically understand PPAR. Additionally, the role of online shopping self-affirmation and regret coping strategies are explored as means to combat PPAR formation. Managerial and theoretical implications are offered, as well as research limitations and future research directions.

No description available.

No description available.

No description available.