Coarticulation: an approach for generating concurrent plans in Markov decision processes Export

@inproceedings{citeulike:1693167, abstract = {We study an approach for performing concurrent activities in Markov decision processes (MDPs) based on the coarticulation framework. We assume that the agent has multiple degrees of freedom (DOF) in the action space which enables it to perform activities simultaneously. We demonstrate that one natural way for generating concurrency in the system is by coarticulating among the set of learned activities available to the agent. In general due to the multiple DOF in the system, often there exists a redundant set of admissible sub-optimal policies associated with each learned activity. Such flexibility enables the agent to concurrently commit to several subgoals according to their priority levels, given a new task defined in terms of a set of prioritized subgoals. We present efficient approximate algorithms for computing such policies and for generating concurrent plans. We also evaluate our approach in a simulated domain.}, address = {New York, NY, USA}, author = {Rohanimanesh, Khashayar and Mahadevan, Sridhar}, booktitle = {ICML '05: Proceedings of the 22nd international conference on Machine learning}, citeulike-article-id = {1693167}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1102442}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1102351.1102442}, comment = {L'articolo propone un'approccio toerico al problema della coordinazione fra azioni differenti per il raggiungimento di obiettivi concorrenti di priorit\`{a} differente. Un algoritmo approssimato per la soluzione del problema viene proposto e alcuni risultati incoraggianti su un problema discreto.}, doi = {10.1145/1102351.1102442}, isbn = {1595931805}, keywords = {algorithm, coarticulation, concurrent, learning, reinforcement}, pages = {720--727}, posted-at = {2007-09-25 15:52:21}, priority = {0}, publisher = {ACM Press}, title = {Coarticulation: an approach for generating concurrent plans in Markov decision processes}, url = {http://dx.doi.org/10.1145/1102351.1102442}, year = {2005} }

TY - CONF ID - citeulike:1693167 L3 - citeulike-article-id:1693167 N2 - We study an approach for performing concurrent activities in Markov decision processes (MDPs) based on the coarticulation framework. We assume that the agent has multiple degrees of freedom (DOF) in the action space which enables it to perform activities simultaneously. We demonstrate that one natural way for generating concurrency in the system is by coarticulating among the set of learned activities available to the agent. In general due to the multiple DOF in the system, often there exists a redundant set of admissible sub-optimal policies associated with each learned activity. Such flexibility enables the agent to concurrently commit to several subgoals according to their priority levels, given a new task defined in terms of a set of prioritized subgoals. We present efficient approximate algorithms for computing such policies and for generating concurrent plans. We also evaluate our approach in a simulated domain. SN - 1595931805 AD - New York, NY, USA EP - 727 TI - Coarticulation: an approach for generating concurrent plans in Markov decision processes PB - ACM Press T2 - ICML '05: Proceedings of the 22nd international conference on Machine learning SP - 720 KW - algorithm KW - coarticulation KW - concurrent KW - learning KW - reinforcement N1 - L'articolo propone un'approccio toerico al problema della coordinazione fra azioni differenti per il raggiungimento di obiettivi concorrenti di priorità differente. Un algoritmo approssimato per la soluzione del problema viene proposto e alcuni risultati incoraggianti su un problema discreto. AU - Rohanimanesh, Khashayar AU - Mahadevan, Sridhar PY - 2005/// UR - http://dx.doi.org/10.1145/1102351.1102442 ER -