2026-04-13T03:58:16Zhttps://recercat.cat/oai/request

oai:recercat.cat:10230/47952025-12-20T01:55:43Zcom_2072_6col_2072_452953

00925njm 22002777a 4500 dc Schlag, Karl author Zapechelnyuk, Andriy author 2017-07-26T10:50:34Z 2017-07-26T10:50:34Z 2009-06-01 2017-07-23T02:12:36Z We consider an agent who has to repeatedly make choices in an uncertain and changing environment, who has full information of the past, who discounts future payoffs, but who has no prior. We provide a learning algorithm that performs almost as well as the best of a given finite number of experts or benchmark strategies and does so at any point in time, provided the agent is sufficiently patient. The key is to find the appropriate degree of forgetting distant past. Standard learning algorithms that treat recent and distant past equally do not have the sequential epsilon optimality property. adaptive learning experts distribution-free e-optimality hannan regret Microeconomics Statistics, Econometrics and Quantitative Methods Decision making in uncertain and changing environments