Strong representation theorems for bitone sequential decision processes
This paper describes a discrete decision process (dds) and a sequential decision process (sdp) by using the language of automata theory. An sdp strongly represents a dds if their sets of feasible policies coincide and cost functions are equal. This paper also considers a special class of an sdp, the so-called bitone sequential decision process (bsdp) and its subclasses: a monotone sequential decision process (msdp), strictly sequential decision process (sbsdp), multiplicative process (mp), and loop-free bitone sequential decision process (lbsdp). The author provides sufficient conditions when for a given dds there exists an sbsdp that strongly represents it. In addition, sufficient conditions are provided for the existence of an mp and for the existence of an lbsdp that strongly represent the given ddp.
- scientific article; zbMATH DE number 6103381
- scientific article; zbMATH DE number 5910584
- STRONG REPRESENTATION OF A DISCRETE DECISION PROCESS BY POSITIVELY/NEGATIVELY BITONE SEQUENTIAL DECISION PROCESS
- Super-strong representation theorems for nondeterministic sequential decision processes
- A representation theorem for a decision theory with conditionals
- Axioms for a Class of Algorithms of Sequential Decision Making
- Strong 1-optimal stationary policies in denumerable Markov decision processes
- The existence of strongly computable representations in the class of Boolean algebras
- Bisimulations and logical characterizations on continuous-time Markov decision processes
- Decision problems, complexity, traces, and representations
This page was built for publication: Strong representation theorems for bitone sequential decision processes
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4664049)