Publication View

Learning mixtures of product distributions over discrete domains (2005)

Abstract
We consider the problem of learning mixtures of product distributions over discrete domains in the distribution learning framework introduced by Kearns et al. We give a $\poly(n/\eps)$ time algorithm for learning a mixture of $k$ arbitrary product distributions over the $n$-dimensional Boolean cube $\{0,1\}^n$ to accuracy $\eps$, for any constant $k$. Previous polynomial time algorithms could only achieve this for $k = 2$ product distributions; our result answers an open question stated independently by Cryan and by Freund and Mansour. We further give evidence that no polynomial time algorithm can succeed when $k$ is superconstant, by reduction from a notorious open problem in PAC learning. Finally, we generalize our $\poly(n/\eps)$ time algorithm to learn any mixture of $k = O(1)$ product distributions over $\{0,1, \dots, b\}^n$, for any $b = O(1)$.

Publication details
Download http://hdl.handle.net/10022/AC:P:29411
Repository Academic Commons by Columbia University Libraries ()