Speaking

2020

• Jan 19 - Hunting deepfakes at DLD, Munich. [video]
• Jan 14 - Deepfakes: commodification, consequences and countermeasures at McKinsey & Company, New York City.

2019

• Dec 14 - Deepfakes: commodification, consequences and countermeasures at NeurIPS19’s workshop on machine learning for creativity and design, Vancouver. [video]
• Nov 19 - The state of deepfakes 2019 at the Global Center for Security Policy, Geneva. [video]
• Nov 19 - The state of deepfakes 2019 at the United Nations, Geneva.
• OCt 19 - Deepfakes: commodification, consequences and countermeasures at the Global forum on AI for humanity, Paris.
• Sep 11 - Trust in time of fakes at DMEXCO, Cologne. [video]
• Aug 26 - Deepfakes, not based on a true story at The Conference, Malmö. [video]
• Jun 19 - Betaworks demo day, New York City.
• Jun 11 - Deepfakes and disinformation at CogX, London. [video]
• Mar 18 - Zeroth demo day, Tokyo.
• Feb 5 - Zeroth demo day, Hong Kong.

2018

• Oct 30 - Sinkhorn autoencoders at University of Amsterdam, Amsterdam.
• May 23 - Commoditization of digital forgery: consequences and countermeasures at Qualcomm, Amsterdam.

2017

• Dec 12 - Federated learning on vertically partitioned data via entity resolution and homomorphic encryption at University of Amsterdam, Amsterdam.
• Sep 22 - Making deep neural networks robust to label noise: a loss correction approach at University of Amsterdam, Amsterdam.
• Aug 11 - Privacy-preserving entity resolution and logistic regression on encrypted data at ICML, Sydney. [slides]
• Jul 23 - Making deep neural networks robust to label noise: a loss correction approach at CVPR, Honolulu. [video] [slides]
• Jul 6 - Making deep neural networks robust to label noise: a loss correction approach at Dat61 CSIRO, Sydney.

2016

• Aug 4 - Weakly supervised learning via statistical sufficiency at ANU/Data61, Canberra. [slides]
• Jun 20 - Loss factorization, weakly supervised learning and label noise robustness at ICML, New York. [video] [slides]

2015

• Nov 19 - Loss factorization and learning from label proportions at Columbia University, New York.
• Nov 11 - Loss factorization and learning from label proportions at Telecom Paris-Tech, Paris.
• Sep 22 - Learning from aggregates at INRIA Parietal Neurospin, Paris.
• Aug 26 - Learning from aggregates at TU Dortmund, Dortmund. [slides]
• Jul 7 - Rademacher observations, private data and boosting at ICML, Lille. [video] [slides]
• Jun 24 - Learning from aggregates at TU Delft, Delft.
• Jun 8 - Learning from aggregates at Ecole Polytechnique, Paris.

2014

• Dec 12 - (Almost) no label no cry at NIPS, Montreal. [slides]
• Aug 28 - Internship final at Commonwealth Bank of Australia, Sydney.
• Jul 10 - (Almost) no label no cry at NICTA, Sydney.
• Feb 27 - Weakly supervised learning and interpretability at Commonwealth Bank of Australia, Sydney.
• Dec 3 - Propensity score matching at NICTA, Sydney.

2013

• Nov 21 - Propensity score matching at Commonwealth Bank of Australia, Sydney.
• Mar 21 - Weakly supervised learning and interpretability at NICTA, Kioloa (Australia).