Titlebook: Stochastic Systems: Modeling, Identification and Optimization II; Roger J.- B. Wets Book 1976Latest edition Springer-Verlag Berlin Heidelb

Strength

迎合

Monotone optimal policies for Markov decision processes,ally ordered and whose action space is a compact subset of the real line. Knowing that such a policy exists, then the search for an optimal policy within the class of all policies can be restricted to the subclass of monotone policies. We also extend (using a new proof) some characterization theorem

UTTER

Desert

beta-cells

A laurent series for the resolvent of a strongly continuous stochastic semi-group,e corresponding resolvent. We show ..=λ...+σ.(−λ)..., assuming .. is a uniform limit of .., at infinity and .=∫.(..−..)d...This Laurent expansion is of interest in the theory of controlled Markov processes. Suppose (..). is a Markov process having transitions (..) and describing the evolution of som

enlist

A laurent series for the resolvent of a strongly continuous stochastic semi-group, at time . is discounted to a present value of e.. Our result expands this total discounted cost as a Laurent series in the interest rate λ..More details are given for finite state Markov chains and diffusion processes on compact intervals.

轻信

Regenerative Markov decision models,l policies do exist and limit decision rules, as the discount-factor tends to one, of discounted optimal rules are bias or equivalently average-overtaken optimal. Finally, an iteration procedure to compute sensitive discount optimal policies is given.

ONYM

Phonophobia

Density functions for random matrix equations,this paper, a partial differential equation for the density function of the solution to the matrix equation is derived. For a restricted class of systems, the solution to the partial differential equation is reduced to the solution of a set of associated ordinary differential equations. A scalar example is provided.

commune

Two-person nonzero sum stochastic differential games with stopping time,A nonzero sum stochastic differential game is considered, with two payoffs, . where . is one dimensional Brownian motion and τ., τ. are stopping times≤.. The existence of a saddle point follows from the existence of a “strong” solution to a system of two quasi-variational inequalities. The proof of existence of a “weak” solution is outlined.