Preselection and expert advice
نویسندگان
چکیده
منابع مشابه
Preselection and expert advice
We study the effects of preselection on an expert’s incentive to give truthful advice. In a decision environment in which certain decisions yield more precise estimates about the expert’s expertise, a mediocre expert’s advice is biased. We show that this bias can be undone by the introduction of a preselection stage, where the decision maker himself sometimes studies the case, and thereby alter...
متن کاملAdvice-Efficient Prediction with Expert Advice
Yevgeny Seldin Peter Bartlett Koby Crammer Mathematical Sciences School, Queensland University of Technology, Brisbane, QLD, Australia Department of Electical Engineering and Computer Sciences, UC Berkeley, Berkeley, CA, USA Department of Computer Science, University College London, London, UK Department of Statistics, UC Berkeley, Berkeley, CA, USA Department of Electrical Engineering, The Tec...
متن کاملCombining Expert Advice Efficiently
We show how models for prediction with expert advice can be defined concisely and clearly using hidden Markov models (HMMs); standard HMM algorithms can then be used to efficiently calculate, among other things, how the expert predictions should be weighted according to the model. We cast many existing models as HMMs and recover the best known running times in each case. We also describe two ne...
متن کاملLearning with Expert Advice∗
Surveys on inflation forecasts show that expectations combine forwardlooking and backward-looking elements. In our paper we rationalize this finding in an equilibrium framework. We assume two types of agents, one having rational expectations and the other using adaptive learning in his forecasts, ratio of the two types in the population evolve according to their past forecasting performance. Ou...
متن کاملActive Learning with Expert Advice
Conventional learning with expert advice methods assume a learner is always receiving the outcome (e.g., class labels) of every incoming training instance at the end of each trial. In real applications, acquiring the outcome from oracle can be costly or time consuming. In this paper, we address a new problem of active learning with expert advice, where the outcome of an instance is disclosed on...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: International Journal of Game Theory
سال: 2016
ISSN: 0020-7276,1432-1270
DOI: 10.1007/s00182-016-0551-9