Michele Flammini
ABSTRACT
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.
- 1.D. Angluin, L.G.Valiant, Fast probabilistic algorithms for Hamiltonian circuits and matchings, J. of Comp. and Syst. Sci., 18, 1979.Google Scholar
- 2.P.L.Bartlett, R.C.Williamson, Investigating the distribution assumptions in the PAC learning model, Proc. COLT 91, Santa Cruz, Morgan and Kaufman, 1991. Google ScholarDigital Library
- 3.A.Blum, M.Singh, Learning functions of k terms, Proc. COLT 90, Morgan and Kaufman, 1990. Google ScholarDigital Library
- 4.A. Ehrenfeucht, D. Haussler, L. Kearns, L. Valiant, A general lower bound on the number of examples needed for learning. Information and Computation 82, 1989. Google ScholarDigital Library
- 5.T.Hancock, Learning monotone k/l formulas on product distribution, Proc. COLT 91, Santa Cruz, Morgan and Kaufman, 1991. Google ScholarDigital Library
- 6.D.Haussler, L. Kearns, N.Littlestone, M. Warmuth, Equivalence of models for polynomial learnability, Proc. $nd Work. on Computational learning Theory, Morgan &: Kauffman, 1988. Google ScholarDigital Library
- 7.M. Kearns, M. Li, Learning in the presence of malicious errors, Proc. ~Oih A CM Symposium on Theory of Computing, 1988. Google ScholarDigital Library
- 8.M. Kearns, M. Li, L. Pitt, L.G. Valiant, On the learnability of boolean formulae, Proc. 19th ACM Symposium on Theory of Computing, 1987. Google ScholarDigital Library
- 9.M. Kearns, L.G. Valiant, Cryptographic limitations on learning boolean formulae and finite automata, Proc. ~1th ACM Symposium on Theory of Computing, 1987. Google ScholarDigital Library
- 10.L.Ku6era, A.Marchetti-Spaccamela, M.Protasi, On learning bounded DNF formulae under uniform distribution, Information and Computation, to appear; a preliminary version appeared in Proc. 15 th Conference on Automata, Languages and Programming, Lecture Notes in Computer Science, vol. 317, Springer Verlag, 1988. Google ScholarDigital Library
- 11.N.Linial, Y.Mansour, N.Nisan, Constant depth circuits, Fourier transform, and learnability, Proc. 30th IEEE Syrup. on Foundations of Computer Science, 1989.Google ScholarDigital Library
- 12.L. Pitt, L.G. Valiant, Computational limitations on learning from examples, J. ACM, 35, 1988. Google ScholarDigital Library
- 13.L.G. Valiant, A theory of the learnable, Communications ACM, 27, 1984. Google ScholarDigital Library
- 14.K.Verbeurgt, Learning DNF under the uniform distribution in quasi-polynomial time, Proc. COLT 90, Morgan and Kaufman, 1990. Google ScholarDigital Library
Index Terms
- Learning DNF formulae under classes of probability distributions
Recommendations
Certain classes of polynomial expansions and multiplication formulas
The authors first present a class of expansions in a series of Bernoulli polyomials and then show how this general result can be applied to yield various (known or new) polynomial expansions. The corresponding expansion problem involving the Euler ...
Exact learning of DNF formulas using DNF hypotheses
Special issue on COLT 2002We show the following: (a) For any @e>0, log^(^3^+^@e^)n-term DNF cannot be polynomial-query learned with membership and strongly proper equivalence queries. (b) For sufficiently large t, t-term DNF formulas cannot be polynomial-query learned with ...
Pseudorandomness for read-k DNF formulas
SODA '19: Proceedings of the Thirtieth Annual ACM-SIAM Symposium on Discrete AlgorithmsThe design of pseudorandom generators and deterministic approximate counting algorithms for DNF formulas are important challenges in unconditional derandomization. Numerous works on these problems have focused on the subclass of small-read DNF formulas, ...
Comments