Density estimation is concerned with the estimation of probability masses, univariate densities, joint densities, and conditional densities. Most of the existing estimators assume that all the data instances are available at once. With the ever increasing amounts of data and a tendency towards online settings, however, there is an increasing demand for density estimation on data streams (LINK). In our group, we particularly work on density estimation of data streams with discrete attributes, continuous attributes, and any mixture thereof. Furthermore, we use the resulting estimate to address applications such as inference, pattern mining, or outlier detection.