Stochastic Shortest Path Problem on Borel Space Considering Cost Distribution

First, we define a stochastic shortest path problem (SSP) on Borel space considering the distribution of total cost. Next, for this SSP, policy classes are set number steps to reach terminal state first time, and is designed each class. There no guarantee that probability large cost small with expected value (expected cost). The minimizes or Conditional Value at Risk (CVaR) may not state. proposed in paper by method reducing δ-perturbed (δ-PP) combining policies obtained discounted problems has low guaranteed Finally, compare methods CVaR.