Random walks and an O*(n5) volume algorithm for convex bodies
Random Structures & Algorithms
Being Bayesian about Network Structure
UAI '00 Proceedings of the 16th Conference on Uncertainty in Artificial Intelligence
Bayesian haplo-type inference via the dirichlet process
ICML '04 Proceedings of the twenty-first international conference on Machine learning
Bayesian hierarchical clustering
ICML '05 Proceedings of the 22nd international conference on Machine learning
Collapsed variational Dirichlet process mixture models
IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
Structured generative models for unsupervised named-entity clustering
NAACL '09 Proceedings of Human Language Technologies: The 2009 Annual Conference of the North American Chapter of the Association for Computational Linguistics
Identification of MCMC samples for clustering
LKR'08 Proceedings of the 3rd international conference on Large-scale knowledge resources: construction and application
Quantum annealing for clustering
UAI '09 Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence
HLT '10 Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics
Distance Dependent Chinese Restaurant Processes
The Journal of Machine Learning Research
Hi-index | 0.00 |
We introduce a new inference algorithm for Dirichlet process mixture models. While Gibbs sampling and variational methods focus on local moves, the new algorithm makes more global moves. This is done by introducing a permutation of the data points as an auxiliary variable. The algorithm is a blocked sampler which alternates between sampling the clustering and sampling the permutation. The key to the efficiency of this approach is that it is possible to use dynamic programming to consider all exponentially many clusterings consistent with a given permutation. We also show that random projections can be used to effectively sample the permutation. The result is a stochastic hill-climbing algorithm that yields burn-in times significantly smaller than those of collapsed Gibbs sampling.