1: \begin{abstract}

2: The three-parameter Indian buffet process is generalized. The possibly

3: different role played by customers is taken into account by suitable

4: (random) weights. Various limit theorems are also proved for such

5: generalized Indian buffet process. Let $L_n$ be the number of dishes

6: experimented by the first $n$ customers, and let $\overline{K}_n=(1/n)

7: \sum_{i=1}^nK_i$ where $K_i$ is the number of dishes tried by customer

8: $i$. The asymptotic distributions of $L_n$ and $\overline{K}_n$,

9: suitably centered and scaled, are obtained. The convergence turns out

10: to be stable (and not only in distribution). As a particular case, the

11: results apply to the standard (i.e., nongeneralized) Indian buffet process.

12: \end{abstract}