scholarly journals The Expected Total Cost Criterion for Markov Decision Processes under Constraints: A Convex Analytic Approach

2012 ◽  
Vol 44 (3) ◽  
pp. 774-793 ◽  
Author(s):  
François Dufour ◽  
M. Horiguchi ◽  
A. B. Piunovskiy
Keyword(s):  
Optimal Control ◽  
Markov Decision Processes ◽  
Optimal Solution ◽  
Decision Processes ◽  
Linear Program ◽  
Occupation Measure ◽  
Stationary Policy ◽  
Total Cost ◽  
Markov Decision ◽  
Expected Total Cost

This paper deals with discrete-time Markov decision processes (MDPs) under constraints where all the objectives have the same form of expected total cost over the infinite time horizon. The existence of an optimal control policy is discussed by using the convex analytic approach. We work under the assumptions that the state and action spaces are general Borel spaces, and that the model is nonnegative, semicontinuous, and there exists an admissible solution with finite cost for the associated linear program. It is worth noting that, in contrast to the classical results in the literature, our hypotheses do not require the MDP to be transient or absorbing. Our first result ensures the existence of an optimal solution to the linear program given by an occupation measure of the process generated by a randomized stationary policy. Moreover, it is shown that this randomized stationary policy provides an optimal solution to this Markov control problem. As a consequence, these results imply that the set of randomized stationary policies is a sufficient set for this optimal control problem. Finally, our last main result states that all optimal solutions of the linear program coincide on a special set with an optimal occupation measure generated by a randomized stationary policy. Several examples are presented to illustrate some theoretical issues and the possible applications of the results developed in the paper.

scholarly journals The Expected Total Cost Criterion for Markov Decision Processes under Constraints: A Convex Analytic Approach

2012 ◽  
Vol 44 (03) ◽  
pp. 774-793 ◽  
Author(s):  
François Dufour ◽  
M. Horiguchi ◽  
A. B. Piunovskiy
Keyword(s):  
Optimal Control ◽  
Markov Decision Processes ◽  
Optimal Solution ◽  
Decision Processes ◽  
Linear Program ◽  
Occupation Measure ◽  
Stationary Policy ◽  
Total Cost ◽  
Markov Decision ◽  
Expected Total Cost

This paper deals with discrete-time Markov decision processes (MDPs) under constraints where all the objectives have the same form of expected total cost over the infinite time horizon. The existence of an optimal control policy is discussed by using the convex analytic approach. We work under the assumptions that the state and action spaces are general Borel spaces, and that the model is nonnegative, semicontinuous, and there exists an admissible solution with finite cost for the associated linear program. It is worth noting that, in contrast to the classical results in the literature, our hypotheses do not require the MDP to be transient or absorbing. Our first result ensures the existence of an optimal solution to the linear program given by an occupation measure of the process generated by a randomized stationary policy. Moreover, it is shown that this randomized stationary policy provides an optimal solution to this Markov control problem. As a consequence, these results imply that the set of randomized stationary policies is a sufficient set for this optimal control problem. Finally, our last main result states that all optimal solutions of the linear program coincide on a special set with an optimal occupation measure generated by a randomized stationary policy. Several examples are presented to illustrate some theoretical issues and the possible applications of the results developed in the paper.

scholarly journals The Expected Total Cost Criterion for Markov Decision Processes under Constraints

2013 ◽  
Vol 45 (3) ◽  
pp. 837-859 ◽  
Author(s):  
François Dufour ◽  
A. B. Piunovskiy
Keyword(s):  
Markov Decision Processes ◽  
Optimal Solution ◽  
Decision Processes ◽  
Linear Program ◽  
Programming Approach ◽  
Stationary Policy ◽  
Total Cost ◽  
Optimal Value ◽  
Markov Decision ◽  
Expected Total Cost

In this work, we study discrete-time Markov decision processes (MDPs) with constraints when all the objectives have the same form of expected total cost over the infinite time horizon. Our objective is to analyze this problem by using the linear programming approach. Under some technical hypotheses, it is shown that if there exists an optimal solution for the associated linear program then there exists a randomized stationary policy which is optimal for the MDP, and that the optimal value of the linear program coincides with the optimal value of the constrained control problem. A second important result states that the set of randomized stationary policies provides a sufficient set for solving this MDP. It is important to note that, in contrast with the classical results of the literature, we do not assume the MDP to be transient or absorbing. More importantly, we do not impose the cost functions to be nonnegative or to be bounded below. Several examples are presented to illustrate our results.

scholarly journals The Expected Total Cost Criterion for Markov Decision Processes under Constraints

2013 ◽  
Vol 45 (03) ◽  
pp. 837-859 ◽  
Author(s):  
François Dufour ◽  
A. B. Piunovskiy
Keyword(s):  
Markov Decision Processes ◽  
Optimal Solution ◽  
Decision Processes ◽  
Linear Program ◽  
Programming Approach ◽  
Stationary Policy ◽  
Total Cost ◽  
Optimal Value ◽  
Markov Decision ◽  
Expected Total Cost

In this work, we study discrete-time Markov decision processes (MDPs) with constraints when all the objectives have the same form of expected total cost over the infinite time horizon. Our objective is to analyze this problem by using the linear programming approach. Under some technical hypotheses, it is shown that if there exists an optimal solution for the associated linear program then there exists a randomized stationary policy which is optimal for the MDP, and that the optimal value of the linear program coincides with the optimal value of the constrained control problem. A second important result states that the set of randomized stationary policies provides a sufficient set for solving this MDP. It is important to note that, in contrast with the classical results of the literature, we do not assume the MDP to be transient or absorbing. More importantly, we do not impose the cost functions to be nonnegative or to be bounded below. Several examples are presented to illustrate our results.

scholarly journals Temporal concatenation for Markov decision processes

2021 ◽  
pp. 1-28
Author(s):  
Ruiyang Song ◽  
Kuang Xu
Keyword(s):  
Markov Decision Processes ◽  
Large Scale ◽  
Optimal Solution ◽  
Upper Bounds ◽  
Black Box ◽  
Decision Processes ◽  
Optimal Solutions ◽  
Wide Range ◽  
Markov Decision ◽  
Speed Up

We propose and analyze a temporal concatenation heuristic for solving large-scale finite-horizon Markov decision processes (MDP), which divides the MDP into smaller sub-problems along the time horizon and generates an overall solution by simply concatenating the optimal solutions from these sub-problems. As a “black box” architecture, temporal concatenation works with a wide range of existing MDP algorithms. Our main results characterize the regret of temporal concatenation compared to the optimal solution. We provide upper bounds for general MDP instances, as well as a family of MDP instances in which the upper bounds are shown to be tight. Together, our results demonstrate temporal concatenation's potential of substantial speed-up at the expense of some performance degradation.

scholarly journals Impulsive Control for Continuous-Time Markov Decision Processes

2015 ◽  
Vol 47 (1) ◽  
pp. 106-127 ◽  
Author(s):  
François Dufour ◽  
Alexei B. Piunovskiy
Keyword(s):  
Optimal Control ◽  
Control Problem ◽  
Markov Decision Processes ◽  
Control Strategy ◽  
Continuous Time ◽  
Sufficient Conditions ◽  
Decision Processes ◽  
Optimal Control Strategy ◽  
Optimality Equation ◽  
Markov Decision

In this paper our objective is to study continuous-time Markov decision processes on a general Borel state space with both impulsive and continuous controls for the infinite time horizon discounted cost. The continuous-time controlled process is shown to be nonexplosive under appropriate hypotheses. The so-called Bellman equation associated to this control problem is studied. Sufficient conditions ensuring the existence and the uniqueness of a bounded measurable solution to this optimality equation are provided. Moreover, it is shown that the value function of the optimization problem under consideration satisfies this optimality equation. Sufficient conditions are also presented to ensure on the one hand the existence of an optimal control strategy, and on the other hand the existence of a ε-optimal control strategy. The decomposition of the state space into two disjoint subsets is exhibited where, roughly speaking, one should apply a gradual action or an impulsive action correspondingly to obtain an optimal or ε-optimal strategy. An interesting consequence of our previous results is as follows: the set of strategies that allow interventions at time t = 0 and only immediately after natural jumps is a sufficient set for the control problem under consideration.

scholarly journals New discount and average optimality conditions for continuous-time Markov decision processes

2010 ◽  
Vol 42 (04) ◽  
pp. 953-985 ◽  
Author(s):  
Xianping Guo ◽  
Liuer Ye
Keyword(s):  
Markov Decision Processes ◽  
Continuous Time ◽  
Average Cost ◽  
Nonnegative Solution ◽  
Decision Processes ◽  
Stationary Policy ◽  
Discounted Cost ◽  
Markov Decision ◽  
Optimal Stationary Policy ◽  
Bounded Below

This paper deals with continuous-time Markov decision processes in Polish spaces, under the discounted and average cost criteria. All underlying Markov processes are determined by given transition rates which are allowed to be unbounded, and the costs are assumed to be bounded below. By introducing an occupation measure of a randomized Markov policy and analyzing properties of occupation measures, we first show that the family of all randomized stationary policies is ‘sufficient’ within the class of all randomized Markov policies. Then, under the semicontinuity and compactness conditions, we prove the existence of a discounted cost optimal stationary policy by providing a value iteration technique. Moreover, by developing a new average cost, minimum nonnegative solution method, we prove the existence of an average cost optimal stationary policy under some reasonably mild conditions. Finally, we use some examples to illustrate applications of our results. Except that the costs are assumed to be bounded below, the conditions for the existence of discounted cost (or average cost) optimal policies are much weaker than those in the previous literature, and the minimum nonnegative solution approach is new.

scholarly journals Impulsive Control for Continuous-Time Markov Decision Processes

2015 ◽  
Vol 47 (01) ◽  
pp. 106-127 ◽  
Author(s):  
François Dufour ◽  
Alexei B. Piunovskiy
Keyword(s):  
Optimal Control ◽  
Control Problem ◽  
Markov Decision Processes ◽  
Control Strategy ◽  
Continuous Time ◽  
Sufficient Conditions ◽  
Decision Processes ◽  
Optimal Control Strategy ◽  
Optimality Equation ◽  
Markov Decision

In this paper our objective is to study continuous-time Markov decision processes on a general Borel state space with both impulsive and continuous controls for the infinite time horizon discounted cost. The continuous-time controlled process is shown to be nonexplosive under appropriate hypotheses. The so-called Bellman equation associated to this control problem is studied. Sufficient conditions ensuring the existence and the uniqueness of a bounded measurable solution to this optimality equation are provided. Moreover, it is shown that the value function of the optimization problem under consideration satisfies this optimality equation. Sufficient conditions are also presented to ensure on the one hand the existence of an optimal control strategy, and on the other hand the existence of a ε-optimal control strategy. The decomposition of the state space into two disjoint subsets is exhibited where, roughly speaking, one should apply a gradual action or an impulsive action correspondingly to obtain an optimal or ε-optimal strategy. An interesting consequence of our previous results is as follows: the set of strategies that allow interventions at time t = 0 and only immediately after natural jumps is a sufficient set for the control problem under consideration.

Sign in / Sign up

Export Citation Format

Share Document