Communications of the ACM
Learning decision trees using the Fourier spectrum
SIAM Journal on Computing
Cryptographic limitations on learning Boolean formulae and finite automata
Journal of the ACM (JACM)
Boosting a weak learning algorithm by majority
Information and Computation
A decision-theoretic generalization of on-line learning and an application to boosting
Journal of Computer and System Sciences - Special issue: 26th annual ACM symposium on the theory of computing & STOC'94, May 23–25, 1994, and second annual Europe an conference on computational learning theory (EuroCOLT'95), March 13–15, 1995
An efficient membership-query algorithm for learning DNF with respect to the uniform distribution
Journal of Computer and System Sciences
More efficient PAC-learning of DNF with membership queries under the uniform distribution
COLT '99 Proceedings of the twelfth annual conference on Computational learning theory
FOCS '99 Proceedings of the 40th Annual Symposium on Foundations of Computer Science
Optimally-Smooth Adaptive Boosting and Application to Agnostic Learning
ALT '02 Proceedings of the 13th International Conference on Algorithmic Learning Theory
| Hi-index | 0.00 |
In the paper, we construct a framework which allows to bound polynomially the distributions produced by certain boosting algorithms, without significant performance loss. Further, we study the case of Freund and Schapire's AdaBoost algorithm, bounding its distributions to near-polynomial w.r.t. the example oracle's distribution. An advantage of AdaBoost over other boosting techniques is that it doesn't require an a-priori accuracy lower bound for the hypotheses accepted from the weak learner during the learning process. We turn AdaBoost into an on-line boosting algorithm (boosting "by filtering"), which can be applied to the wider range of learning problems. In particular, now AdaBoost applies to the problem of DNF-learning, answering affirmatively the question posed by Jackson. We also construct a hybrid boosting algorithm, in that way achieving the lowest bound possible for booster-produced distributions (in terms of Õ), and show a possible application to the problem of DNF-learning w.r.t. the uniform.