BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//wp-events-plugin.com//7.2.3.1//EN
BEGIN:VEVENT
UID:62@lincs.fr
DTSTART;TZID=Europe/Paris:20160226T140000
DTEND;TZID=Europe/Paris:20160226T150000
DTSTAMP:20170313T170920Z
URL:https://www.lincs.fr/events/on-the-whittle-index-for-restless-multi-ar
 med-hidden-markov-bandits/
SUMMARY:On the Whittle Index for Restless Multi-armed Hidden Markov Bandits
DESCRIPTION:\n\n\nAbstract:\nWe consider a restless multi-armed bandit in
 which each armcan be in one two states. When an arm is sampled\, the state
 of the armis not available to the sampler. Instead\, a binary signal with a
 knownrandomness that depends on the state of the arm is made available.
 Nosignal is displayed if the arm is not sampled. An arm-dependent reward
 isaccrued from each sampling. In each time step\, each arm changes
 stateaccording to known transition probabilities which in turn depend
 onwhether the arm is sampled or not sampled. Since the state of the arm
 isnever visible and has to be inferred from the current belief and
 apossible binary signal\, we call this the hidden Markov bandit.
 Ourinterest is in a policy to select the arm(s) in each time step
 thatmaximises the infinite horizon discounted reward. Specifically\, we
 seekthe use of Whittle’s index in selecting the arms.We
 first analyze the single-armed bandit and show that it admits anapproximate
 threshold-type optimal policy when the
 ‘no-sample’ actionis subsidised. Next\, we
 show that this also satisfies an approximate-indexability property.
 Numerical examples support the analytical results\n\n\n\n&nbsp\;
CATEGORIES:Seminars
LOCATION:LINCS Meeting Room 40\, 23\, avenue d'Italie\, Paris\, 75013\,
 France
GEO:48.8283983;2.3568972000000485
X-APPLE-STRUCTURED-LOCATION;VALUE=URI;X-ADDRESS=23\, avenue d'Italie\,
 Paris\, 75013\, France;X-APPLE-RADIUS=100;X-TITLE=LINCS Meeting Room
 40:geo:48.8283983,2.3568972000000485
END:VEVENT
BEGIN:VTIMEZONE
TZID:Europe/Paris
X-LIC-LOCATION:Europe/Paris
BEGIN:STANDARD
DTSTART:20151025T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
END:STANDARD
END:VTIMEZONE
END:VCALENDAR