KALE: When Energy-Based Learning Meets Adversarial Training

Mar 10, 2020
Michael Arbel, Liang Zhou, Arthur Gretton

  Access Model/Code and Paper

Learning Deep Kernels for Non-Parametric Two-Sample Tests

Feb 21, 2020
Feng Liu, Wenkai Xu, Jie Lu, Guangquan Zhang, Arthur Gretton, D. J. Sutherland

  Access Model/Code and Paper

A kernel log-rank test of independence for right-censored data

Dec 08, 2019
Tamara Fernandez, Arthur Gretton, David Rindt, Dino Sejdinovic

  Access Model/Code and Paper

Kernelized Wasserstein Natural Gradient

Oct 25, 2019
Michael Arbel, Arthur Gretton, Wuchen Li, Guido Montufar

  Access Model/Code and Paper

Counterfactual Distribution Regression for Structured Inference

Aug 20, 2019
Nicolo Colombo, Ricardo Silva, Soong M Kang, Arthur Gretton

* 24 pages, 5 figures 

  Access Model/Code and Paper

A Kernel Stein Test for Comparing Latent Variable Models

Jul 01, 2019
Heishiro Kanagawa, Wittawat Jitkrittum, Lester Mackey, Kenji Fukumizu, Arthur Gretton

  Access Model/Code and Paper

Maximum Mean Discrepancy Gradient Flow

Jun 11, 2019
Michael Arbel, Anna Korba, Adil Salim, Arthur Gretton

  Access Model/Code and Paper

Kernel Instrumental Variable Regression

Jun 01, 2019
Rahul Singh, Maneesh Sahani, Arthur Gretton

* 31 pages, 8 figures 

  Access Model/Code and Paper

Exponential Family Estimation via Adversarial Dynamics Embedding

Apr 27, 2019
Bo Dai, Zhen Liu, Hanjun Dai, Niao He, Arthur Gretton, Le Song, Dale Schuurmans

* 66 figures, 25 pages; preliminary version published in NeurIPS2018 Bayesian Deep Learning Workshop 

  Access Model/Code and Paper

Learning deep kernels for exponential family densities

Nov 22, 2018
Li Wenliang, Dougal Sutherland, Heiko Strathmann, Arthur Gretton

  Access Model/Code and Paper

Kernel Exponential Family Estimation via Doubly Dual Embedding

Nov 06, 2018
Bo Dai, Hanjun Dai, Arthur Gretton, Le Song, Dale Schuurmans, Niao He

* 22 pages, 20 figures 

  Access Model/Code and Paper

On gradient regularizers for MMD GANs

Oct 27, 2018
Michael Arbel, Dougal J. Sutherland, Mikołaj Bińkowski, Arthur Gretton

* Code available at https://github.com/MichaelArbel/Scaled-MMD-GAN . v2: NIPS camera-ready version 

  Access Model/Code and Paper

Informative Features for Model Comparison

Oct 27, 2018
Wittawat Jitkrittum, Heishiro Kanagawa, Patsorn Sangkloy, James Hays, Bernhard Schölkopf, Arthur Gretton

* Accepted to NIPS 2018 

  Access Model/Code and Paper

BRUNO: A Deep Recurrent Model for Exchangeable Data

Oct 16, 2018
Iryna Korshunova, Jonas Degrave, Ferenc Huszár, Yarin Gal, Arthur Gretton, Joni Dambre

* NIPS 2018 

  Access Model/Code and Paper

Antithetic and Monte Carlo kernel estimators for partial rankings

Jul 25, 2018
Maria Lomeli, Mark Rowland, Arthur Gretton, Zoubin Ghahramani

  Access Model/Code and Paper

Kernel Conditional Exponential Family

Apr 08, 2018
Michael Arbel, Arthur Gretton

  Access Model/Code and Paper

Model-based Kernel Sum Rule: Kernel Bayesian Inference with Probabilistic Models

Apr 05, 2018
Yu Nishiyama, Motonobu Kanagawa, Arthur Gretton, Kenji Fukumizu

  Access Model/Code and Paper

Demystifying MMD GANs

Mar 21, 2018
Mikołaj Bińkowski, Dougal J. Sutherland, Michael Arbel, Arthur Gretton

* Published at ICLR 2018: https://openreview.net/forum?id=r1lUOzWCW . v4: actually-final version: non-existence of unbiased estimators for IPMs and FID; clarity edits to the main proof 

  Access Model/Code and Paper

Efficient and principled score estimation with Nyström kernel exponential families

Mar 13, 2018
Dougal J. Sutherland, Heiko Strathmann, Michael Arbel, Arthur Gretton

* v5: Final version to be published at AISTATS 2018 

  Access Model/Code and Paper

A Linear-Time Kernel Goodness-of-Fit Test

Oct 24, 2017
Wittawat Jitkrittum, Wenkai Xu, Zoltan Szabo, Kenji Fukumizu, Arthur Gretton

* Accepted to NIPS 2017 

  Access Model/Code and Paper

Density Estimation in Infinite Dimensional Exponential Families

May 26, 2017
Bharath Sriperumbudur, Kenji Fukumizu, Arthur Gretton, Aapo Hyvärinen, Revant Kumar

* 58 pages, 8 figures; Fixed some errors and typos 

  Access Model/Code and Paper

Generative Models and Model Criticism via Optimized Maximum Mean Discrepancy

Feb 10, 2017
Dougal J. Sutherland, Hsiao-Yu Tung, Heiko Strathmann, Soumyajit De, Aaditya Ramdas, Alex Smola, Arthur Gretton

* Published at ICLR 2017 (public comments: http://openreview.net/forum?id=HJWHIKqgl ). v4: minor edits 

  Access Model/Code and Paper

Fast Non-Parametric Tests of Relative Dependency and Similarity

Nov 17, 2016
Wacha Bounliphone, Eugene Belilovsky, Arthur Tenenhaus, Ioannis Antonoglou, Arthur Gretton, Matthew B. Blashcko

  Access Model/Code and Paper

Interpretable Distribution Features with Maximum Testing Power

Oct 28, 2016
Wittawat Jitkrittum, Zoltan Szabo, Kacper Chwialkowski, Arthur Gretton

  Access Model/Code and Paper

Learning Theory for Distribution Regression

Oct 21, 2016
Zoltan Szabo, Bharath Sriperumbudur, Barnabas Poczos, Arthur Gretton

* Journal of Machine Learning Research, 17(152):1-40, 2016 
* Final version appeared at JMLR, with supplement. Code: https://bitbucket.org/szzoli/ite/. arXiv admin note: text overlap with arXiv:1402.1754 

  Access Model/Code and Paper

An Adaptive Test of Independence with Analytic Kernel Embeddings

Oct 15, 2016
Wittawat Jitkrittum, Zoltan Szabo, Arthur Gretton

* 8 pages of main text 

  Access Model/Code and Paper

Recovery of non-linear cause-effect relationships from linearly mixed neuroimaging data

Sep 30, 2016
Sebastian Weichwald, Arthur Gretton, Bernhard Schölkopf, Moritz Grosse-Wentrup

* Pattern Recognition in Neuroimaging (PRNI), International Workshop on, 1-4, 2016 
* arXiv admin note: text overlap with arXiv:1512.01255 

  Access Model/Code and Paper

MERLiN: Mixture Effect Recovery in Linear Networks

Sep 27, 2016
Sebastian Weichwald, Moritz Grosse-Wentrup, Arthur Gretton

* IEEE Journal of Selected Topics in Signal Processing, 10(7), 1254-1266, 2016 

  Access Model/Code and Paper

A Kernel Test of Goodness of Fit

Sep 27, 2016
Kacper Chwialkowski, Heiko Strathmann, Arthur Gretton

* 14 pages, 9 figures 

  Access Model/Code and Paper