121
Voted
APPROX
2008
Springer
15 years 4 months ago
2008
Springer
We give an algorithm that with high probability properly learns random monotone DNF with t(n) terms of length log t(n) under the uniform distribution on the Boolean cube {0, 1}n ....