Clustering mixture models in almost-linear time via list-decodable mean estimation.
Ilias DiakonikolasDaniel M. KaneDaniel KongsgaardJerry LiKevin TianPublished in: STOC (2022)
Keyphrases
- mixture model
- density estimation
- unsupervised learning
- mixture modeling
- model based clustering
- gaussian mixture model
- overlapping clustering
- em algorithm
- finite mixture models
- generative model
- probabilistic model
- clustering algorithm
- bayesian information criterion
- k means
- probability density
- probabilistic mixture model
- expectation maximization
- finite mixtures
- language model
- clustering method
- maximum likelihood
- hierarchical clustering
- probability density function
- density function
- model selection
- gaussian mixture
- maximum likelihood estimation
- document clustering
- supervised learning
- finite mixture model
- learning algorithm
- parameter estimation
- mixture components
- object recognition
- minimum message length
- computer vision
- data sets