论文信息 - Learning DNF formulae under classes of probability distributions

Learning DNF formulae under classes of probability distributions

We show that 2-term DNF formulae are learnable in quadratic time using only a logarithmic number of positive examples if we assume that examples are drawn from a bounded distribution. We also show that k-term DNF formulae are learnable in polynomial time using positive and negative examples drawn from a bounded distribution.

Michele Flammini | Alberto Marchetti-Spaccamela | Ludek Kucera

[1] Ming Li,et al. Learning in the presence of malicious errors , 1993, STOC '88.

[2] David Haussler,et al. Equivalence of models for polynomial learnability , 1988, COLT '88.

[3] Yishay Mansour,et al. Learning monotone ku DNF formulas on product distributions , 1991, COLT '91.

[4] Leslie G. Valiant,et al. On the learnability of Boolean formulae , 1987, STOC.

[5] Noam Nisan,et al. Constant depth circuits, Fourier transform, and learnability , 1993, JACM.

[6] Leslie G. Valiant,et al. Computational limitations on learning from examples , 1988, JACM.

[7] Karsten A. Verbeurgt. Learning DNF under the uniform distribution in quasi-polynomial time , 1990, COLT '90.

[8] Leslie G. Valiant,et al. A general lower bound on the number of examples needed for learning , 1988, COLT '88.

[9] Mona Singh,et al. Learning functions of k terms , 1990, COLT '90.

[10] Leslie G. Valiant,et al. A theory of the learnable , 1984, CACM.

[11] Leslie G. Valiant,et al. Fast probabilistic algorithms for hamiltonian circuits and matchings , 1977, STOC '77.

[12] Leslie G. Valiant,et al. Cryptographic Limitations on Learning Boolean Formulae and Finite Automata , 1993, Machine Learning: From Theory to Applications.

[13] Peter L. Bartlett,et al. Investigating the distribution assumptions in the Pac learning model , 1991, COLT '91.