Essentially all data mining algorithms assume that the datagenerating process is independent of the data miner's activities. However, in many domains, including spam detectio...
Nilesh N. Dalvi, Pedro Domingos, Mausam, Sumit K. ...
We develop a semi-supervised learning method that constrains the posterior distribution of latent variables under a generative model to satisfy a rich set of feature expectation c...
Traditional feature selection methods assume that the data are independent and identically distributed (i.i.d.). In real world, tremendous amounts of data are distributed in a net...
We develop nonparametric Bayesian models for multiscale representations of images depicting natural scene categories. Individual features or wavelet coefficients are marginally de...
Jyri J. Kivinen, Erik B. Sudderth, Michael I. Jord...