Linear unit-tests for invariance discovery
Benjamin Aubin, Agnieszka Słowik, Martin Arjovsky, Léon Bottou, David Lopez-Paz
People
David Lopez-Paz
Research Scientist
I’m a research scientist at Facebook AI Research, Paris, France. Prior to joining Facebook in 2016, I received my PhD from the Max Planck Institute for Intelligent Systems and the University of Cambridge, advised by Bernhard Schölkopf and Zoubin Ghahramani. Before that, I obtained a degree in computer science from the Universidad Autonoma de Madrid in 2011. During my studies, I interned at various industries, including the European Space Agency, Google Research and the Red Bull Formula 1 team.
The goal of my research is to develop theory and algorithms for unsupervised causal inference, and their use to build machines which are able to reason and learn about the world using less data.
Interests
Learning theory, causal inference, nonconvex optimization, un/semi/supervised learning and new learning paradigms
Latest Publications
Causal Discovery & Causality Workshop at NeurIPS - 2021
Permutation Equivariant Models for Compositional Generalization in Language
Jonathan Gordon, David Lopez-Paz, Marco Baroni, Diane Bouchacourt
First-order Adversarial Vulnerability of Neural Networks and Input Dimension
Carl-Johann Simon-Gabriel, Yann Ollivier, Bernhard Scholkopf, Léon Bottou, David Lopez-Paz
Manifold Mixup: Better Representations by Interpolating Hidden States
Vikas Verma, Alex Lamb, Christopher Beckham, Amir Najafi, Ioannis Mitliagkas, David Lopez-Paz, Yoshua Bengio
Geometrical Insights for Implicit Generative Modeling
Léon Bottou, Martin Arjovsky, David Lopez-Paz, Maxime Oquab
Optimizing the Latent Space of Generative Networks
Piotr Bojanowski, Armand Joulin, David Lopez-Paz, Arthur Szlam
