发帖
雷达卡
Non-Negative Bregman Divergence Minimization
for Deep Direct Density Ratio Estimation
Masahiro Kato 1 Takeshi Teshima 2
Abstract the density ratio when anomaly-free unlabeled test data are
available (Hido et al., 2008).
Density ratio estimation (DRE) is at the core of
various machine learning tasks such as anomaly Among the various approaches to DRE, we focus on the
detection and d ...
京公网安备 11010802022788号
