Abstract

We show that the class of monotone 2 O( √ log n)-term DNF formulae can be PAC learned in polynomial time under the uniform distribution from random examples only. This is an exponential improvement over the best previous polynomial-time algorithms in this model, which could learn monotone o(log 2 n)-term DNF. We also show that various classes of small constant-depth circuits which compute monotone functions are PAC learnable in polynomial time under the uniform distribution. All of our results extend to learning under any constant-bounded product distribution.

Citations

1515	A theory of the learnable - Valiant - 1984
583	Queries and concept learning - Angluin - 1988
253	Constant Depth Circuits, Fourier Transform, and Learnability - Linial, Mansour, et al. - 1993
190	A guided tour of Chernoff bounds - Hagerup, Rub - 1990
190	The influence of variables on Boolean functions - Kahn, Kalai, et al. - 1988
168	The complexity of finite functions - Boppana, Sipser - 1992
163	Computational Limitations for Small Depth Circuits - H˚astad - 1987
155	On the Learnability of Boolean Formula - Kearns, Li, et al. - 1987
150	An efficient membership-query algorithm for learning dnf with respect to the uniform distribution - Jackson - 1994
105	Weakly learning dnf and characterizing statistical query learning using fourier analysis - Blum, Furst, et al. - 1994
82	Exact Learning via the Monotone Theory - Bshouty - 1993
47	Fast Learning of k-term DNF formulas with Queries - Blum, Rudich - 1992
46	A switching lemma primer - Beame - 1994
44	A representation of the joint distribution of responses to n dichotomous items - Bahadur - 1961
44	Exact learning boolean functions via the monotone theory - Bshouty - 1995
44	Learning DNF under the uniform distribution in quasi-polynomial time - Verbeurgt - 1990
42	On the Fourier Spectrum of Monotone Functions - Bshouty, Tamon - 1996
38	An O(n log log n ) learning algorithm for DNF under the uniform distribution - Mansour - 1995
36	Boosting and Hardcore Sets - Klivans, Servedio - 1999
36	A guided tour of Cherno bounds - Hagerup, Rub - 1990
33	Monotone versus positive - Ajtai, Gurevich - 1987
31	More efficient PAC-learning of DNF with membership queries under the uniform distribution - Bshouty, Jackson, et al.
24	Simple learning algorithms using divide and conquer - BSHOUTY - 1995
23	Improved learning of ac 0 functions - Furst, Jackson, et al. - 1991
22	On using the Fourier transform to learn disjoint DNF - Khardon - 1994
22	Learning sub-classes of monotone DNF on the uniform distribution - Verbeurgt - 1998
20	Learning monotone log-term DNF formulas under the uniform distribution - Sakai, Maruoka
17	Improved learning of AC ° functions - Furst, Jackson, et al. - 1991
16	A technique for upper bounding the spectral norm with applications to learning - BELLARE - 1992
16	On learning monotone DNF formulae under uniform distributions - Kucera, Marchetti-Spaccamela, et al. - 1994
13	Learning monotone k-� DNF formulas on product distributions - HANCOCK, MANSOUR - 1991
10	Learning monotone k-µ DNF formulas on product distributions - Hancock, Mansour - 1991
7	The complexity of learning formulas and decision trees that have restricted reads - Hancock - 1992
7	The complexity of functions - Boppana, Sipser - 1990
7	Queries and concept learning. Machine Learning 2(4):319–342. 8.3.1 - Angluin - 1988
6	Learning monotone k-u DNF formulas on product distributions - Hancock, Mansour - 1991
5	More e#cient PAC learning of DNF with membership queries under the uniform distribution - Bshouty, Jackson, et al. - 1999
2	On the applicationf of multiplicity automata - Beimel, Bergadano, et al. - 1996
1	On the influence of negations on the complexity of a realization of monotone Boolean functions by formulas of bounded depth, Metody Diskret - Okol'nishnikova - 1982
1	A simple algorithm for learing O(log n)-term DNF - Kushilevitz - 1997