BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:Detecting radiological anomalies - James Scott (Un
iversity of Texas at Austin)
DTSTART;TZID=Europe/London:20170706T110000
DTEND;TZID=Europe/London:20170706T114500
UID:TALK73174AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/73174
DESCRIPTION:Radiologically active materials are used widely in
industry\, medicine\, and research. Yet an unsecu
red\, lost\, or stolen radiological source can pre
sent a major threat to public safety.  \;To de
al with the potential environmental and security h
azards posed by such a scenario\, govenment agenci
es use various detection procedures at ports of en
try to their countries.  \;Moreover\, security
agencies that try to prevent terrorist attacks ar
e keenly interested in the problem of identifying
and locating stolen or smuggled radiation samples.
Even at the local level\, police departments have
shown increasing interest in the deployment of sy
stems for detecting anomalous radiological sources
.
Statistically speaking\, the radiological
anomaly-detection problem is one of detecting a c
hange in distribution. Sequential data is collecte
d from a sensor that measures the energies of arri
ving gamma rays. These observed energies are rando
m variables drawn from an energy spectrum\, which
is a probability distribution over the set of poss
ible gamma-ray energies. The question is whether t
hose measured energies are from the normal backgro
und spectrum\, and therefore harmless\, or whether
they are from an anomalous spectrum due to the pr
esence of a nearby radiological source.  \;In
this talk I will describe some new statistical met
hods we&rsquo\;ve developed for deal with two majo
r challenges in this setting: 1) characterizing th
e spatially varying background radiation in dense
urban areas\; and 2) flagging anomalous readings f
rom spatially distributed sensor networks in a sta
tistically rigorous way.
This is joint work
with Wesley Tansey\, Oscar Padilla\, Alex Reinhar
t\, and Alex Athey.
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR