The strength of weak learnability

@article{citeulike:2858063, abstract = {This paper addresses the problem of improving the accuracy of an hypothesis output by a learning algorithm in the distribution-free (PAC) learning model. A concept class islearnable (orstrongly learnable) if, given access to a source of examples of the unknown concept, the learner with high probability is able to output an hypothesis that is correct on all but an arbitrarily small fraction of the instances. The concept class isweakly learnable if the learner can produce an hypothesis that performs only slightly better than random guessing. In this paper, it is shown that these two notions of learnability are equivalent.}, author = {Schapire, Robert E. }, citeulike-article-id = {2858063}, doi = {10.1007/BF00116037}, journal = {Machine Learning}, keywords = {algorithms, complexity, kdd}, month = {June}, number = {2}, pages = {197--227}, posted-at = {2008-06-03 03:08:47}, priority = {2}, title = {The strength of weak learnability}, url = {http://dx.doi.org/10.1007/BF00116037}, volume = {5}, year = {1990} }

TY - JOUR ID - citeulike:2858063 L3 - citeulike-article-id:2858063 TI - The strength of weak learnability JF - Machine Learning VL - 5 IS - 2 SP - 197 EP - 227 N2 - This paper addresses the problem of improving the accuracy of an hypothesis output by a learning algorithm in the distribution-free (PAC) learning model. A concept class islearnable (orstrongly learnable) if, given access to a source of examples of the unknown concept, the learner with high probability is able to output an hypothesis that is correct on all but an arbitrarily small fraction of the instances. The concept class isweakly learnable if the learner can produce an hypothesis that performs only slightly better than random guessing. In this paper, it is shown that these two notions of learnability are equivalent. KW - algorithms KW - complexity KW - kdd AU - Schapire, Robert PY - 1990/06/01/ UR - http://dx.doi.org/10.1007/BF00116037 ER -