Corpus ID: 2449064

Regularization Strategies and Empirical Bayesian Learning for MKL

@article{Tomioka2010RegularizationSA,
  title={Regularization Strategies and Empirical Bayesian Learning for MKL},
  author={Ryota Tomioka and Taiji Suzuki},
  journal={ArXiv},
  year={2010},
  volume={abs/1011.3090}
}
  • Ryota Tomioka, Taiji Suzuki
  • Published 2010
  • Mathematics, Computer Science
  • ArXiv
  • Multiple kernel learning (MKL), structured sparsity, and multi-task learning have recently received considerable attention. In this paper, we show how different MKL algorithms can be understood as applications of either regularization on the kernel weights or block-norm-based regularization, which is more common in structured sparsity and multi-task learning. We show that these two regularization strategies can be systematically mapped to each other through a concave conjugate operation. When… CONTINUE READING

    Citations

    Publications citing this paper.
    SHOWING 1-10 OF 15 CITATIONS

    Generalized Multiple Kernel Learning With Data-Dependent Priors

    VIEW 1 EXCERPT
    CITES METHODS

    Fast Learning Rate of lp-MKL and its Minimax Optimality

    VIEW 1 EXCERPT
    CITES BACKGROUND

    Regularizers for structured sparsity

    VIEW 1 EXCERPT
    CITES METHODS

    PAC-Bayesian Bound for Gaussian Process Regression and Multiple Kernel Additive Model

    VIEW 2 EXCERPTS
    CITES METHODS & BACKGROUND

    Structured sparsity with convex penalty functions

    VIEW 1 EXCERPT
    CITES METHODS

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 56 REFERENCES