Efficient important sampling for a feed-forward network
Seminar Room 1, Newton Institute
AbstractWe consider a feedforward network with a single server station serving two classes of jobs, where one class of jobs has preemptive priority over the other. The rare event of interest is total population overflow. We rigorously identify the large deviation rate of the rare event probabilities and construct piecewise constant, asymptotically optimal importance sampling schemes.
If it doesn't, something may have gone wrong with our embedded player.
We'll get it fixed as soon as possible.