User profiles for C. J. Simon-Gabriel
Carl-Johann Simon-GabrielAmazon Web Services - Tübingen Lablet Verified email at tue.mpg.de Cited by 1102 |
Adagan: Boosting generative models
…, O Bousquet, CJ Simon-Gabriel… - Advances in neural …, 2017 - proceedings.neurips.cc
Generative Adversarial Networks (GAN) are an effective method for training generative models
of complex data such as natural images. However, they are notoriously hard to train and …
of complex data such as natural images. However, they are notoriously hard to train and …
Assaying out-of-distribution generalization in transfer learning
…, P Gehler, CJ Simon-Gabriel… - Advances in …, 2022 - proceedings.neurips.cc
Since out-of-distribution generalization is a generally ill-posed problem, various proxy targets
(eg, calibration, adversarial robustness, algorithmic corruptions, invariance across shifts) …
(eg, calibration, adversarial robustness, algorithmic corruptions, invariance across shifts) …
Bridging the gap to real-world object-centric learning
Humans naturally decompose their environment into entities at the appropriate level of
abstraction to act in the world. Allowing machine learning algorithms to derive this decomposition …
abstraction to act in the world. Allowing machine learning algorithms to derive this decomposition …
From optimal transport to generative modeling: the VEGAN cookbook
We study unsupervised generative modeling in terms of the optimal transport (OT) problem
between true (but unknown) data distribution $P_X$ and the latent variable model …
between true (but unknown) data distribution $P_X$ and the latent variable model …
First-order adversarial vulnerability of neural networks and input dimension
CJ Simon-Gabriel, Y Ollivier, L Bottou… - International …, 2019 - proceedings.mlr.press
Over the past few years, neural networks were proven vulnerable to adversarial images:
targeted but imperceptible image perturbations lead to drastically different predictions. We show …
targeted but imperceptible image perturbations lead to drastically different predictions. We show …
Kernel distribution embeddings: Universal kernels, characteristic kernels and kernel metrics on distributions
CJ Simon-Gabriel, B Schölkopf - Journal of Machine Learning Research, 2018 - jmlr.org
Kernel mean embeddings have become a popular tool in machine learning. They map
probability measures to functions in a reproducing kernel Hilbert space. The distance between …
probability measures to functions in a reproducing kernel Hilbert space. The distance between …
Modeling confounding by half-sibling regression
…, D Janzing, CJ Simon-Gabriel… - Proceedings of the …, 2016 - National Acad Sciences
We describe a method for removing the effect of confounders to reconstruct a latent quantity
of interest. The method, referred to as “half-sibling regression,” is inspired by recent work in …
of interest. The method, referred to as “half-sibling regression,” is inspired by recent work in …
Metrizing weak convergence with maximum mean discrepancies
… explicitly point out the flaw in the proof of Claim 2 by Simon-Gabriel and Schölkopf (2018). …
The flaw in the proof of Theorem 12 of Simon-Gabriel and Schölkopf (2018) (our Claim 2) …
The flaw in the proof of Theorem 12 of Simon-Gabriel and Schölkopf (2018) (our Claim 2) …
Adversarial vulnerability of neural networks increases with input dimension
Over the past four years, neural networks have been proven vulnerable to adversarial images:
targeted but imperceptible image perturbations lead to drastically different predictions. We …
targeted but imperceptible image perturbations lead to drastically different predictions. We …
Targeted separation and convergence with kernel discrepancies
Kernel Stein discrepancies (KSDs) are maximum mean discrepancies (MMDs) that leverage
the score information of distributions, and have grown central to a wide range of applications…
the score information of distributions, and have grown central to a wide range of applications…