Final published version
Research output: Contribution to Journal/Magazine › Journal article › peer-review
Research output: Contribution to Journal/Magazine › Journal article › peer-review
}
TY - JOUR
T1 - A characterization of product-form exchangeable feature probability functions
AU - Battiston, Marco
AU - Favaro, Stefano
AU - Roy, Daniel M.
AU - Teh, Yee Whye
PY - 2018/6/1
Y1 - 2018/6/1
N2 - We characterize the class of exchangeable feature allocations assigning probability Vn,k∏kl=1WmlUn−ml to a feature allocation of n individuals, displaying k features with counts (m1,…,mk) for these features. Each element of this class is parametrized by a countable matrix V and two sequences U and W of nonnegative weights. Moreover, a consistency condition is imposed to guarantee that the distribution for feature allocations of (n−1) individuals is recovered from that of n individuals, when the last individual is integrated out. We prove that the only members of this class satisfying the consistency condition are mixtures of three-parameter Indian buffet Processes over the mass parameter γ, mixtures of N-dimensional Beta–Bernoulli models over the dimension N, or degenerate limits thereof. Hence, we provide a characterization of these two models as the only consistent exchangeable feature allocations having the required product form, up to randomization of the parameters.
AB - We characterize the class of exchangeable feature allocations assigning probability Vn,k∏kl=1WmlUn−ml to a feature allocation of n individuals, displaying k features with counts (m1,…,mk) for these features. Each element of this class is parametrized by a countable matrix V and two sequences U and W of nonnegative weights. Moreover, a consistency condition is imposed to guarantee that the distribution for feature allocations of (n−1) individuals is recovered from that of n individuals, when the last individual is integrated out. We prove that the only members of this class satisfying the consistency condition are mixtures of three-parameter Indian buffet Processes over the mass parameter γ, mixtures of N-dimensional Beta–Bernoulli models over the dimension N, or degenerate limits thereof. Hence, we provide a characterization of these two models as the only consistent exchangeable feature allocations having the required product form, up to randomization of the parameters.
KW - Indian buffet process
KW - exchangeable feature allocations
KW - Gibbs-type partitions
U2 - 10.1214/17-AAP1333
DO - 10.1214/17-AAP1333
M3 - Journal article
VL - 28
SP - 1423
EP - 1448
JO - Annals of Applied Probability
JF - Annals of Applied Probability
SN - 1050-5164
IS - 3
ER -