About me

I am a fourth year Ph.D. student at Gatsby under the supervision of Arthur Gretton. Prior to joining Gatsby, I graduated from Ecole polytechnique with a focus in Applied Mathematics and got a Masters Degree in Mathematics, Machine Learning and Computer Vision (MVA). I also worked as a Computer Vision Engineer at Prophesee where I developed tracking algorithms based on signals from neuromorophic cameras. I am currently interested in unsupervised learning and distributional optimization with an emphasis on GANs and Generalized Natural Gradients. More details can be found in my resume.

News

01 / 2020
12 / 2019
ICLR2020 paper accepted with a spotlight talk on Kernelized Wasserstein Natural Gradient
12 / 2019
Poster presentation at NeurIPS2019 on Maximum Mean Discrepancy Gradient Flow (Vancouver, Canada)
07 / 2019
12 / 2018
Poster presentation at NeurIPS2018 on On gradient regularizers for MMD GANs (Montreal, Canada)
10 / 2018
Poster presentation at Microsoft Research PhD Summit 2018 (Redmond, USA)
07 / 2018
Best Poster Award at MSR AI Summer School 2018 (Cambridge, UK)
06 / 2018
Poster presentation at Data Science Summer School 2018 (Palaiseau, France)
06 / 2018
Code released for Scaled MMD-GAN.
04 / 2018
Poster presentation at AISTATS2018 on Kernel Conditional Exponential Family (Lanzarote, Spain)
03 / 2018
02 / 2018
Attended the Workshop on Functional Inference and Machine Learning 2018, Institute of Statistical Mathematics (Tachikawa, Japan)
09 / 2017
Data Science Summer School 2017 (Palaiseau, France)
09 / 2017
Attended the Scalable statistical inference workshop 2017, Isaac Newton Institute (Cambridge, UK)

Publications

KALE: When Energy-Based Learning Meets Adversarial Training
Michael Arbel, Liang Zhou, Arthur Gretton.
Under review
Synchronizing Probability Measures on Rotations via Optimal Transport
Tolga Birdal, Michael Arbel, Umut Simsekli, Leonidas Guibas.
CVPR, 2020.
Kernelized Wasserstein Natural Gradient
Michael Arbel, Arthur Gretton, Wuchen Li, Guido Montufar.
ICLR, 2020.
Maximum Mean Discrepancy Gradient Flow
Michael Arbel, Anna Korba, Adil Salim and Arthur Gretton.
NeurIPS, 2019.
On gradient regularizers for MMD GANs
Michael Arbel*, Dougal J. Sutherland*, Mikołaj Bińkowski and Arthur Gretton.
NeurIPS, 2018.
Demystifying MMD GANs
Mikołaj Bińkowski*, Dougal J. Sutherland*, Michael Arbel and Arthur Gretton.
ICLR, 2018
Kernel Conditional Exponential Family
Michael Arbel and Arthur Gretton.
AISTATS, 2018
Efficient and principled score estimation with Nyström kernel exponential families
Dougal J. Sutherland*, Heiko Strathmann*, Michael Arbel and Arthur Gretton.
AISTATS, 2018

Invited talks

02 / 2020
02 / 2020
Department of Statistics, University of Oxford (Oxford, UK).
11 / 2019
The Alan Turing Institute (London, UK).
09 / 2019
Amazon Research Days (Berlin, Germany).
09 / 2019
03 / 2019
11 / 2018
Cambridge-Tübingen workshop (Tenerife, Spain).
11 / 2018
Google Developer Group Reading & Thames Valley (Reading, UK)

Activities

Fall 2017-2019
Fall 2017
Fall 2017