Neural Information Processing Systems (NIPS) 2012, Lake Tahoe, NV (Workshop on Modern Nonparametric Methods in Machine Learning). "Posterior consistency for the number of components in a finite mixture."

Correction: There was an error in the proof of claims (B) and (C) of this abstract. (Claim (B) was that inconsistency remains when a prior is placed on the concentration parameter, and claim (C) was that inconsistency remains when the concentration parameter is modified to prevent the number of clusters from diverging.) We conjecture that these claims are true, but we do not have a proof. Claims (A) and (D) are correct --- there is inconsistency when the concentration parameter is fixed. Also, the results presented in our papers on this topic (at NIPS 2013 and JMLR 2014) are correct.