a variant of the stochastic shortest path (SSP) problem

Hey all,

As we know, there is a detonation/absorbing state in SSP problem and
we can reach the destination with probability one under certain
policies. The problem is to derive an optimal policy so as to minimize
the total expected cost until reaching the absorbing state.

Right now, I am considering a variant of SSP: the number of horizons
is fixed. Thereby, we may reach the absorbing state within the
deadline or we can not. The goal is to minimize the total expected
cost until reaching the absorbing state or the deadline, subject to
the probability of reaching the absorbing state is larger than a pre-
specified value. I am wondering whether there exists any reference
close to the above problem.

Really appreciate any of your comment.

coolzai

.

Quantcast