Nonmyopic active learning of Gaussian processes: an exploration-exploitation approach Export

@inproceedings{citeulike:1741522, abstract = {When monitoring spatial phenomena, such as the ecological condition of a river, deciding where to make observations is a challenging task. In these settings, a fundamental question is when an active learning, or sequential design, strategy, where locations are selected based on previous measurements, will perform significantly better than sensing at an a priori specified set of locations. For Gaussian Processes (GPs), which often accurately model spatial phenomena, we present an analysis and efficient algorithms that address this question. Central to our analysis is a theoretical bound which quantifies the performance difference between active and a priori design strategies. We consider GPs with unknown kernel parameters and present a nonmyopic approach for trading off exploration, i.e., decreasing uncertainty about the model parameters, and exploitation, i.e., near-optimally selecting observations when the parameters are (approximately) known. We discuss several exploration strategies, and present logarithmic sample complexity bounds for the exploration phase. We then extend our algorithm to handle nonstationary GPs exploiting local structure in the model. We also present extensive empirical evaluation on several real-world problems.}, address = {New York, NY, USA}, author = {Krause, Andreas and Guestrin, Carlos}, booktitle = {ICML '07: Proceedings of the 24th international conference on Machine learning}, citeulike-article-id = {1741522}, citeulike-linkout-0 = {http://portal.acm.org/citation.cfm?id=1273553}, citeulike-linkout-1 = {http://dx.doi.org/10.1145/1273496.1273553}, doi = {10.1145/1273496.1273553}, isbn = {978-1-59593-793-3}, keywords = {active, active\_learning, gaussian, learning, processes}, location = {Corvalis, Oregon}, pages = {449--456}, posted-at = {2008-01-25 17:05:40}, priority = {4}, publisher = {ACM}, title = {Nonmyopic active learning of Gaussian processes: an exploration-exploitation approach}, url = {http://dx.doi.org/10.1145/1273496.1273553}, year = {2007} }

TY - CONF ID - citeulike:1741522 L3 - citeulike-article-id:1741522 N2 - When monitoring spatial phenomena, such as the ecological condition of a river, deciding where to make observations is a challenging task. In these settings, a fundamental question is when an active learning, or sequential design, strategy, where locations are selected based on previous measurements, will perform significantly better than sensing at an a priori specified set of locations. For Gaussian Processes (GPs), which often accurately model spatial phenomena, we present an analysis and efficient algorithms that address this question. Central to our analysis is a theoretical bound which quantifies the performance difference between active and a priori design strategies. We consider GPs with unknown kernel parameters and present a nonmyopic approach for trading off exploration, i.e., decreasing uncertainty about the model parameters, and exploitation, i.e., near-optimally selecting observations when the parameters are (approximately) known. We discuss several exploration strategies, and present logarithmic sample complexity bounds for the exploration phase. We then extend our algorithm to handle nonstationary GPs exploiting local structure in the model. We also present extensive empirical evaluation on several real-world problems. CY - Corvalis, Oregon SN - 978-1-59593-793-3 AD - New York, NY, USA EP - 456 TI - Nonmyopic active learning of Gaussian processes: an exploration-exploitation approach PB - ACM T2 - ICML '07: Proceedings of the 24th international conference on Machine learning SP - 449 KW - active KW - active_learning KW - gaussian KW - learning KW - processes AU - Krause, Andreas AU - Guestrin, Carlos PY - 2007/// UR - http://dx.doi.org/10.1145/1273496.1273553 ER -