S. Zilberstein and A-I. Mouaddib. Reactive control of dynamic progressive processing. In IJCAI, pages 1268--1273, 1999.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....are also exploring other forms of on board plan modification. We are collaborating on a project that is exploring Markov model approaches to selecting among possible task decompositions, evaluating goal achievement, and potentially reordering goals to optimize plan quality (Bernstein et al. 2001; Zilberstein Mouaddib 1999). This will allow more extensive plan modifications, while respecting the original plan goals. Acknowledgments: We would like to acknowledge David E. Smith, Keith Golden, and Trey Smith for their contributions to the design of the Contingent Rover Language. We would also like to acknowledge the ....

Zilberstein, S., and Mouaddib, A.-I. 1999. Reactive control of dynamic progressive processing. In Proceedings of IJCAI, 1268--1273.

....on the dynamic operation, scalability is also enhanced by the proposed solution. Our approach is to exploit the fact that the units of the plan are largely independent. We try to capture the dependency of the execution of each unit on the remaining plan using a notion similar to opportunity cost [Zilberstein and Mouaddib, 1999] . 4.1 Dynamic control with one resource Definition 13 Let ##### represent the optimal local value of the state after step with remaining resources . Note that # is the same as for a modified plan in which unit # has no successors. # is simply the value of the best policy for PRU ....

S. Zilberstein and A-I. Mouaddib. Reactive control of dynamic progressive processing. In IJCAI, pages 1268--1273, 1999.

....Their dynamic programming approach provides a globally optimal schedule. Hansen and Zilberstein have developed similar policies to control interruptible anytime algorithms [8] More recently, the approach has been applied successfully to control a complex progressive pro cessing task structure [22]. These examples demonstrate that MDPs can be used effectively to handle uncertainty in computation and develop optimal meta level control policies. 7. Conclusion We have analyzed the problem of optimal sequencing of contract algorithms. The problem arises when there is uncertainty about the ....

Shlomo Zilberstein and Abdel-Illah Mouaddib. Reactive control of dynamic progressive processing. Sixteenth International Joint Conference on Artificial Intelligence, 1268-1273, 1999.

....of the plan. The dynamic information includes the remaining workload and the remaining resources, both of which can be captured by the notion of opportunity cost. Each plan assigned to a rover is composed of a sequence of target activities represented as progressive processing task structures [24, 35]. An initial resource allocation is also specified. Resources are represented as vectors of discrete units. We assume here that the plan is totally ordered and that resources are not renewable. A generalization of the technique to acyclic graphs has been examined in [10] The rover can perform a ....

Zilberstein, S., Mouaddib, A.-I.: Reactive Control of Dynamic Progressive Processing. Sixteenth International Joint Conference on Artificial Intelligence (1999) 1268--1273

....Recognizing that a non myopic approach can lead to better performance, Harada and Russell [11] consider treating this meta level control problem as a Markov decision problem and propose using reinforcement learning with value function approximation techniques to solve it. Mouaddib and Zilberstein [23,41] have developed a similar framework for controlling progressive processing task structures. In this model, a system satisfies a set of information requests under time pressure by limiting the amount of processing allocated to each task based on a predefined hierarchical task structure. Using ....

S. Zilberstein, A.-I. Mouaddib, Reactive control of dynamic progressive processing, in: Proc. IJCAI-99, Stockholm, Sweden, 1999, pp. 1268--1273.

....of the plan. The dynamic information includes the remaining workload and the remaining resources, both of which can be captured by the notion of opportunity cost. Each plan assigned to a rover is composed of a sequence of target activities represented as progressive processing task structures [21, 30]. An initial resource allocation is also specified. Resources are represented as vectors of discrete units. We assume here that the plan is totally ordered and that resources are not renewable. A generalization of the technique to acyclic graphs has been examined in [9] 2.1 The Rover Model The ....

S. Zilberstein and A.I. Mouaddib. Reactive control of dynamic progressive processing. Sixteenth International Joint Conference on Artificial Intelligence, 1268--1273, 1999.

No context found.

Shlomo Zilberstein and Abdel-Illah Mouaddib. Reactive control of dynamic progressive processing. IJCAI, pp. 1268--1273, 1999.

No context found.

Shlomo Zilberstein and Abdel-Illah Mouaddib. Reactive control of dynamic progressive processing. In Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence (IJCAI), pages 1268-- 1273, Stockholm, Sweden, 1999.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC