Corpus ID: 49397352

Deep Orthogonal Representations: Fundamental Properties and Applications

@article{Hsu2018DeepOR,
  title={Deep Orthogonal Representations: Fundamental Properties and Applications},
  author={Hsiang Hsu and Salman Salamatian and Fl{\'a}vio du Pin Calmon},
  journal={ArXiv},
  year={2018},
  volume={abs/1806.08449}
}
  • Hsiang Hsu, Salman Salamatian, Flávio du Pin Calmon
  • Published in ArXiv 2018
  • Computer Science, Mathematics
  • Several representation learning and, more broadly, dimensionality reduction techniques seek to produce representations of the data that are orthogonal (uncorrelated). Examples include PCA, CCA, Kernel/Deep CCA, the ACE algorithm and correspondence analysis (CA). For a fixed data distribution, all finite variance representations belong to the same function space regardless of how they are derived. In this work, we present a theoretical framework for analyzing this function space, and demonstrate… CONTINUE READING

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    11
    Twitter Mentions

    Citations

    Publications citing this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 37 REFERENCES

    Principal Inertia Components and Applications

    VIEW 9 EXCERPTS

    Elements of Information Theory

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Maximally Correlated Principal Component Analysis

    VIEW 11 EXCERPTS
    HIGHLY INFLUENTIAL

    On Deep Multi-View Representation Learning

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    On Sequences of Pairs of Dependent Random Variables

    VIEW 7 EXCERPTS
    HIGHLY INFLUENTIAL

    On measures of dependence

    VIEW 5 EXCERPTS
    HIGHLY INFLUENTIAL

    Analysis of Boolean Functions

    VIEW 1 EXCERPT
    HIGHLY INFLUENTIAL

    Deep Canonical Correlation Analysis

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning Multiple Layers of Features from Tiny Images

    VIEW 6 EXCERPTS
    HIGHLY INFLUENTIAL