Information-theoretic clustering and mixture learning

Clustering

• Robust clustering:
  • total Jensen clustering (ICASSP 2015)
     
    
  • total Bregman clustering, soft and hard, Shape Retrieval Using Hierarchical Total Bregman Soft Clustering (TPAMI 2012)
  
   
  
• Mixed divergence clustering, dual pair of centroids per cluster, On Clustering Histograms with k-Means by Using Mixed a-Divergences (Entropy 2014)
   
  
• Generalization of k-means, Further heuristics for k-means: The merge-and-split heuristic and the (k,l)-means (arxiv 2014)

Learning statistical mixtures

• total Jensen clustering, Total Jensen divergences: Definition, properties and clustering (ICASSP 2015)
   
  
• k-MLE, On learning statistical mixtures maximizing the complete likelihood (MaxEnt 2014)
   
  
• Mixture by simplifying KDE, Learning mixtures by simplifying kernel density estimators (Matrix Information Geometry 2013)
   
  
• Jointly learn set of mixtures, Comix: Joint Estimation and Lightspeed Comparison of Mixture Models (ICASSP 2016)
   
  
• Fisher-Rao clustering of mixtures
   
  
• Hierarchical representation of mixtures, Simplification and hierarchical representations of mixtures of exponential families (Signal Processing, 2010)
   
  
• Software libraries in Java/Python/C: jMEF and PyMEF libMEF

• Bregman hard clustering, Bregman soft clustering:
  Clustering with Bregman Divergences, Arindam Banerjee, Srujana Merugu, Inderjit S. Dhillon, Joydeep Ghosh (JMLR 2015).