Fabrice Daniel

Background

About Fabrice Daniel

Manage a Research department in Machine Learning application to the Business domains covered by the company, including mainly- AI Based automated Trading Strategies - Credit Card Fraud detection - Music Recommender Systems We also built a strong long term relationship with CentraleSupelec and Polytechnique, providing great AI research projects to student teams every year since 2012. In 2019 CentraleSupelec and Lusis have signed off a Research University Chair for 4 years, with 2 PhDs, 3 Professors, 30 students and 5 AI Lusis team to drive research, provide datasets, frameworks and business knowledge. The research we are conducting in this chair are focused on two subjects: 1- Trading strategies : Make predictions on stochastic processes. While the models will be built on financial market predictions, it’s expected to be applicable to any other stochastic processes 2- Credit card fraud : Explainability of Deep Learning models in the context of credit card fraud detection For 2019/2020 we also have 21 students in 4 teams from CentraleSupelec and 3 teams from Polytechnique working on the following subjects- Predict price direction with Convolutional Networks - Learn trading strategy with deep reinforcement learning - Credit card fraud detection with deep learning - Apply graph based machine learning to credit card fraud detection - Explainability of Deep Learning based models for credit card fraud detection - Create a metric reflecting the real expected performances of a market direction predictor - Determine the similarity of two musics from their sound data by using unsupervised approaches In 2020 two interns from Polytechnique are working on AI subjects relative to trading strategies- Use Machine Learning to optimize trade exit inside the prediction horizon of existing AI based trading strategies running in production - Use Machine Learning to optimize a portfolio of existing AI based trading strategies running in production Two other interns from CentraleSupelec and Telecom Nancy are working on AI for music recommender subject- Music similarity from sound spectrograms - Neural collaborative filtering If you have any concern with applied AI in your company, feel free to contact me on LinkedIn.