Recovering Distributions from Gaussian RKHS Embeddings
By Motonobu Kanagawa, Kenji Fukumizu. 2014
34 citations recorded by the source corpus.
Subfields: Representation & Self-Supervised Learning, Probabilistic & Bayesian Methods
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Representation & Self-Supervised Learning
- Bernhard Schölkopf
- Bharath K. Sriperumbudur
- Kenji Fukumizu
- Krikamol Muandet
- Arthur Gretton
- Jan Peters
- Yu Nishiyama
- Motonobu Kanagawa
In Probabilistic & Bayesian Methods
- Bernhard Schölkopf
- Kenji Fukumizu
- Krikamol Muandet
- Bharath K. Sriperumbudur
- Arthur Gretton
- Yu Nishiyama
- Motonobu Kanagawa
- Jan Peters
Citations in the corpus
- Kernel Mean Embedding of Distributions: A Review and Beyonds (2016, background): “We respond to this question by discussing two closely related problems, namely, the distributional pre-image problem11 (Kwok and Tsang 2004, Song et al. 2008, Kanagawa and Fukumizu 2014) and kernel herding (Chen et al. 2010).”
- Computing functions of random variables via reproducing kernel Hilbert space representations (2015, positive): “Alternatively, if p has a density, we can estimate it using methods such as reproducing kernel moment matching and combinations with kernel density estimation (Song et al. 2008; Kanagawa and Fukumizu 2014).”
- Filtering with State-Observation Examples via Kernel Monte Carlo Filter (2013, background): “ion of fcan be estimated by the weighted average P n i=1 w if(X i). Note that here fis a function in the RKHS, but the same can also be shown for functions outside the RKHS under certain assumptions (Kanagawa and Fukumizu, 2014). In this way, the estimator of the form (5) provides estimators of mome”