BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Discriminating Classical and Quantum Channels - Bjarne Bergh\, Uni
 versity of Cambridge
DTSTART:20230301T140000Z
DTEND:20230301T150000Z
UID:TALK197644@talks.cam.ac.uk
CONTACT:Prof. Ramji Venkataramanan
DESCRIPTION:Channel discrimination is in its simplest form is the hypothes
 is testing problem where we are given a black-box channel which can be one
  of two candidate channels and the task is to find out which one it is. We
  are interested in optimal asymptotic error rates for this problem\, and i
 n particular whether adaptivity (i.e. choosing the inputs of the channel u
 ses based on previous outcomes) is required for optimal strategies. For th
 e simple discrimination problem just described\, it has previously been sh
 own that adaptivity is asymptotically not required when the channels are c
 lassical\, whereas in the quantum case adaptivity gives an advantage for t
 he symmetric but not the asymmetric error exponent. We study the more gene
 ral composite problem\, i.e. the problem where we don’t have two single 
 candidate channels\, but two sets of candidate channels\, and focus on the
  asymmetric case. There we show that for classical channels adaptivity can
  give an asymptotic advantage\, however we also prove optimality of non-ad
 aptive strategies when the sets of channels are convex. For the equivalent
  problem with quantum channels we prove an entropic expression for the Ste
 in exponent using non-adaptive strategies. \n\nThe talk will start by intr
 oducing the problem for classical channels\, and illustrate previous work 
 and associated results. It will subsequently give a brief introduction of 
 the required concepts from quantum information theory before addressing th
 e quantum problem. 
LOCATION:MR5\, CMS Pavilion A
END:VEVENT
END:VCALENDAR