论文信息 - Dynamic optimization: a grand unification

Dynamic optimization: a grand unification

It is commonly thought that deterministic control is in principle a special case of stochastic control. It is argued that actually the inclusion is the reverse: without the characteristic nonanticipativity requirement a stochastic problem is equivalent to a family of deterministic problems (one for each omega ), and nonanticipativity can be enforced by bringing in a suitable Lagrange multiplier.<<ETX>>

Mark H. A. Davis

[1] Mark H. A. Davis,et al. Deterministic Methods in Stochastic Optimal Control. , 1992 .

[2] R. Tyrrell Rockafellar,et al. Scenarios and Policy Aggregation in Optimization Under Uncertainty , 1991, Math. Oper. Res..

[3] Mark H. A. Davis,et al. On the Value of Information in Controlled Diffusion Processes , 1991 .

[4] Mtw,et al. Stochastic flows and stochastic differential equations , 1990 .

[5] Mark H. A. Davis. Anticipative LQG Control , 1989 .

[6] Kerry Back,et al. The shadow price of information in continuous time decision problems , 1987 .

[7] M. Dempster. The Expected Value of Perfect Information in the Optimal Evolution of Stochastic Systems , 1981 .

[8] Mark H. A. Davis. Linear estimation and stochastic control , 1977 .

[9] R. Rockafellar,et al. Nonanticipativity and L1-martingales in stochastic optimization problems , 1976 .

[10] Roger J.-B. Wets,et al. On the Relation between Stochastic and Deterministic Optimization , 1975 .