Compressive K-means

@article{Keriven2017CompressiveK,
  title={Compressive K-means},
  author={N. Keriven and N. Tremblay and Y. Traonmilin and R. Gribonval},
  journal={2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
  year={2017},
  pages={6369-6373}
}
  • N. Keriven, N. Tremblay, +1 author R. Gribonval
  • Published 2017
  • Computer Science, Mathematics
  • 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
  • The Lloyd-Max algorithm is a classical approach to perform K-means clustering. Unfortunately, its cost becomes prohibitive as the training dataset grows large. We propose a compressive version of K-means (CKM), that estimates cluster centers from a sketch, i.e. from a drastically compressed representation of the training dataset. We demonstrate empirically that CKM performs similarly to Lloyd-Max, for a sketch size proportional to the number of centroids times the ambient dimension, and… CONTINUE READING
    31 Citations

    Figures and Topics from this paper

    Quantized Compressive K-Means
    • 10
    • Highly Influenced
    • PDF
    Differentially Private Compressive K-means
    • 6
    • PDF
    Large-Scale High-Dimensional Clustering with Fast Sketching
    • 11
    • PDF
    Compressive Learning for Semi-Parametric Models
    • 1
    • PDF
    Sketched clustering via hybrid approximate message passing
    • 7
    • PDF
    Sketching for large-scale learning of mixture models
    • 48
    • PDF
    Approximating Spectral Clustering via Sampling: a Review
    • 12
    • PDF
    Sketched Clustering via Hybrid Approximate Message Passing
    • 2
    • Highly Influenced
    • PDF

    References

    SHOWING 1-10 OF 36 REFERENCES
    Random Projections for $k$-means Clustering
    • 119
    • Highly Influential
    • PDF
    Linear-Time Learning on Distributions with Approximate Kernel Embeddings
    • 15
    • PDF
    Sketching for large-scale learning of mixture models
    • 48
    • PDF
    Compressive Spectral Clustering
    • 78
    • PDF
    A fast k-means implementation using coresets
    • 99
    • PDF
    Unsupervised Feature Selection for the $k$-means Clustering Problem
    • 107
    • PDF
    Efficient Kernel Clustering Using Random Fourier Features
    • 63
    • PDF
    A unified framework for approximating and clustering data
    • 273
    • PDF
    k-means++: the advantages of careful seeding
    • 5,677
    • PDF
    Fast Approximate Nearest Neighbors with Automatic Algorithm Configuration
    • 2,645
    • PDF