Material Detail

This video was recorded at NIPS Workshops, Whistler 2009. Model selection in clustering requires (i) to specify a clustering principle and (ii) to decide an appropriate number of clusters depending on the noise level in the data. We advocate an information theoretic perspective where the uncertainty in the data set induces an uncertainty in the solution space of clusterings. A clustering model, which can tolerate a higher level of noise in the data than competing models, is considered to be superior provided that the clustering solution is equally informative. This tradeoff between informativeness and robustness is used as a model selection criterion. The request that solutions should generalize from one data set to an equally probable second data set gives rise to a new notion of structure induced information.

Disciplines:

• Science and Technology  / Computer Science  / Programming & Programming Languages