Continuous-Time Hierarchial Reinforcement Learning Export

@inproceedings{citeulike:1704523, abstract = {Hierarchical reinforcement learning (RL) is a general framework which studies how to exploit the structure of actions and tasks to accelerate policy learning in large domains. Prior work in hierarchical RL, such as the MAXQ method, has been limited to the discrete-time discounted reward semiMarkov decision process (SMDP) model. This paper generalizes the MAXQ method to continuous-time discounted and average reward SMDP models. We describe two hierarchical reinforcement learning algorithms:}, author = {Ghavamzadeh, Mohammad and Mahadevan, Sridhar}, booktitle = {Proc. 18th International Conf. on Machine Learning}, citeulike-article-id = {1704523}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.28.3413}, keywords = {algorithm, continuous\_time, hierarchical, learning, reinforcement}, pages = {186--193}, posted-at = {2007-09-28 11:16:42}, priority = {3}, publisher = {Morgan Kaufmann, San Francisco, CA}, title = {Continuous-Time Hierarchial Reinforcement Learning}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.28.3413}, year = {2001} }

TY - CONF ID - citeulike:1704523 L3 - citeulike-article-id:1704523 N2 - Hierarchical reinforcement learning (RL) is a general framework which studies how to exploit the structure of actions and tasks to accelerate policy learning in large domains. Prior work in hierarchical RL, such as the MAXQ method, has been limited to the discrete-time discounted reward semiMarkov decision process (SMDP) model. This paper generalizes the MAXQ method to continuous-time discounted and average reward SMDP models. We describe two hierarchical reinforcement learning algorithms: EP - 193 TI - Continuous-Time Hierarchial Reinforcement Learning PB - Morgan Kaufmann, San Francisco, CA T2 - Proc. 18th International Conf. on Machine Learning SP - 186 KW - algorithm KW - continuous_time KW - hierarchical KW - learning KW - reinforcement AU - Ghavamzadeh, Mohammad AU - Mahadevan, Sridhar PY - 2001/// UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.28.3413 ER -