Using kernel perceptrons to learn action effects for planning Export

@inproceedings{Mourao:2008, abstract = {We investigate the problem of learning action effects in STRIPS and ADL planning domains. Our approach is based on a kernel perceptron learning model, where action and state information is encoded in a compact vector representation as input to the learning mechanism, and resulting state changes are produced as output. Empirical results of our approach indicate efficient training and prediction times, with low average error rates (< 3\%) when tested on STRIPS and ADL versions of an object manipulation scenario. This work is part of a project to integrate machine learning techniques with a planning system, as part of a larger cognitive architecture linking a high level reasoning component with a low-level robot/vision system.}, author = {Mour\~{a}o, Kira and Petrick, Ronald P. A. and Steedman, Mark}, booktitle = {Proceedings of the International Conference on Cognitive Systems (CogSys 2008)}, citeulike-article-id = {2717051}, citeulike-linkout-0 = {http://www.anc.ed.ac.uk/dtc/administrator/components/com_publish/files/publications/Mourao2008CogSys.pdf}, keywords = {action-planning, kernel-trick, perceptron-learning-rule, phd, prediction}, month = {April}, posted-at = {2008-04-25 09:41:15}, priority = {0}, title = {Using kernel perceptrons to learn action effects for planning}, url = {http://www.anc.ed.ac.uk/dtc/administrator/components/com_publish/files/publications/Mourao2008CogSys.pdf}, year = {2008} }

TY - CONF ID - Mourao:2008 L3 - citeulike-article-id:2717051 N2 - We investigate the problem of learning action effects in STRIPS and ADL planning domains. Our approach is based on a kernel perceptron learning model, where action and state information is encoded in a compact vector representation as input to the learning mechanism, and resulting state changes are produced as output. Empirical results of our approach indicate efficient training and prediction times, with low average error rates (< 3%) when tested on STRIPS and ADL versions of an object manipulation scenario. This work is part of a project to integrate machine learning techniques with a planning system, as part of a larger cognitive architecture linking a high level reasoning component with a low-level robot/vision system. TI - Using kernel perceptrons to learn action effects for planning T2 - Proceedings of the International Conference on Cognitive Systems (CogSys 2008) KW - action-planning KW - kernel-trick KW - perceptron-learning-rule KW - phd KW - prediction AU - Mourão, Kira AU - Petrick, Ronald AU - Steedman, Mark PY - 2008/04// UR - http://www.anc.ed.ac.uk/dtc/administrator/components/com_publish/files/publications/Mourao2008CogSys.pdf ER -