The following papers have been accepted for presentation at the International Joint Conference on Neural Networks (IJCNN): “On Ensemble Techniques for Data Stream Regression” Authors: Heitor Murilo Gomes, Jacob Montiel, Saulo Martiello Mastelini, Bernhard Pfahringer and Albert Bifet “Adaptive XGBoost for Evolving Data Streams” Authors: Jacob Montiel, Rory Mitchell, Eibe Frank, Bernhard Pfahringer, Talel Abdessalem and Albert Bifet

Our tutorial “Machine learning for data streams with scikit-multiflow” has been accepted for presentation in IJCAI-PRICAI 2020! Abstract Data stream mining has gained a lot of attention in recent years as an exciting research topic. However, there is still a gap between the pure research proposals and the practical applications to real-world machine learning problems. The main goal of this tutorial is to introduce attendees to data stream mining theory and practice.

Our tutorial for stream learning with scikit-multiflow has been accepted for presentation at the International Joint Conference on Neural Networks (IJCNN) to take part alongside the IEEE World Congress on Computational Intelligence (WCCI) 2020.

My interview on the benefits of adaptive learning methods has been published in a white-book by Quantmetry. The white-book, No. 5 “AI en production”, is free to download here. Transcript of the interview (in french) Qu’est-ce qu’une dérive ? L’apprentissage est souvent considéré comme une tâche statique. Cependant, en conditions réelles, les données évoluent constamment. C’est ce qu’on appelle une dérive conceptuelle. Par exemple, les marchés financiers sont instables : les risques de crédit ne sont pas les mêmes d’une année sur l’autre.

I have started a new potion as Research Fellow in the Machine Learning Group at the University of Waikato in New Zealand.

In the field of machine learning on data streams, data is assumed infinite and models are trained and updated continuously, thereby adapting to changes in the data. This talk provides an overview of data stream learning and introduces scikit-multiflow, an open-source Python framework to easily implement algorithms and perform experiments. As traditional “batch” learning struggles to keep in pace with today’s data deluge, a parallel field emerges — data stream mining.

I am participating in the Model life-cycle round-table at DataJob Paris 2019. A great opportunity to talk about stream learning and its relevance on real-world applications. Companies in multiple sectors are actively deploying machine learning models into production, but that is not the end of the story, models need to be monitored and maintained. Furthermore, in multiple real-world applications data changes over time, turning trained models obsolete. Participants: Heloïse Nonne | Head of Data Science & Engineering | ITNOVEM, SNCF expertise center Renan Borne | Chief Architect | AXA-IM John Whitbeck | Head of Machine Learning | Liftoff Grégoire Martinon | Senior Data Scientist | Quantmetry Jacob Montiel | Research fellow | University of Waikato

As traditional “batch” learning is no match for today’s data deluge, a new field emerges — data stream mining. In stream learning, data is considered infinite and models are trained and updated continuously, thereby adapting to changes in the data. This talk provides an overview of data stream learning and introduces scikit-multiflow, an open-source Python framework to implement algorithms and perform experiments in the field of ML on evolving data streams.

Boosting is an ensemble method that combines base models in a sequential manner to achieve high predictive accuracy. A popular learning …

Data ubiquity highlights the need of efficient and adaptable data-driven solutions. In this paper, we present FAST AND SLOW LEARNING …

scikit-multiflow is a framework for learning from data streams and multi-output learning in Python. Conceived to serve as a platform to …

Missing data is a common trait of real-world data that can negatively impact interpretability. In this paper, we present Cascade …

Detecting over-indebtedness, the difficulties meeting household payment commitments, poses multiple Big Data challenges for banking …