Corpus ID: 14671775

Conditional mean embeddings as regressors

@inproceedings{Grnewlder2012ConditionalME,
  title={Conditional mean embeddings as regressors},
  author={Steffen Gr{\"u}new{\"a}lder and G. Lever and A. Gretton and Luca Baldassarre and Sam Patterson and M. Pontil},
  booktitle={ICML},
  year={2012}
}
  • Steffen Grünewälder, G. Lever, +3 authors M. Pontil
  • Published in ICML 2012
  • Mathematics, Computer Science
  • We demonstrate an equivalence between reproducing kernel Hilbert space (RKHS) embeddings of conditional distributions and vector-valued regressors. This connection introduces a natural regularized loss function which the RKHS embeddings minimise, providing an intuitive understanding of the embeddings and a justification for their use. Furthermore, the equivalence allows the application of vector-valued regression methods and results to the problem of learning conditional distributions. Using… CONTINUE READING
    85 Citations
    Active Learning of Conditional Mean Embeddings via Bayesian Optimisation
    • 1
    • Highly Influenced
    • PDF
    Hyperparameter Learning for Conditional Mean Embeddings with Rademacher Complexity Bounds
    Hyperparameter Learning for Conditional Kernel Mean Embeddings with Rademacher Complexity Bounds
    A Rigorous Theory of Conditional Mean Embeddings\ast
    Kernel Distribution Embeddings
    • 2012
    Learning from Conditional Distributions via Dual Kernel Embeddings
    • 25
    • PDF
    A Measure-Theoretic Approach to Kernel Conditional Mean Embeddings
    • 5
    • Highly Influenced
    • PDF
    Bayesian Deconditional Kernel Mean Embeddings
    • 2
    • PDF
    Compressed Conditional Mean Embeddings for Model-Based Reinforcement Learning
    • 16
    • PDF

    References

    SHOWING 1-10 OF 23 REFERENCES
    Modelling transition dynamics in MDPs with RKHS embeddings
    • 86
    • PDF
    Hilbert space embeddings of conditional distributions with applications to dynamical systems
    • 211
    • Highly Influential
    • PDF
    Optimal Rates for the Regularized Least-Squares Algorithm
    • 432
    • PDF
    On Learning Vector-Valued Functions
    • 414
    • PDF
    Universal Multi-Task Kernels
    • 125
    • PDF
    Kernel dimension reduction in regression
    • 242
    • PDF
    A Hilbert Space Embedding for Distributions
    • 577
    • PDF
    Kernel Measures of Conditional Dependence
    • 367
    • PDF
    Gaussian process dynamic programming
    • 193
    • PDF
    Reproducing kernel Hilbert spaces in probability and statistics
    • 1,141
    • Highly Influential
    • PDF