Subhasish Basak

Presentation @ DCE Reading group

10 March 2021

Made a presentation at the Data Centric Engineering (DCE) reading group of The Alan Turing Institute (ATI), London on our article "Numerical issues in maximum likelihood parameter estimation for Gaussian process interpolation ", with Sébastien Petit (Safran Aircraft Engines), Julien Bect and Emmanuel Vazquez (L2S, CentraleSupelec).

Considering the widespread use of Gaussian Processes in the domain of statistics and machine learning, this study provides some simple and effective strategies for robust GP implementations.

• Find more on: DCE Reading Group

• Slides: DOI:10.5281/zenodo.4653845

Keywords: Gaussian Processes, Optimization, Machine learning

Numerical analysis Summer school 2021

Université Pierre-et-Marie CURIE, Paris

14-18 June

Attended summer school entitled "Multi-fidelity, multi-level, model selection/aggregation: how the presence of several versions of a code can improve the prediction of complex phenomena."

• Find more on: Summer School CEA 2021

Keywords: python, model selection, Gaussian Processes

ArtiSaneFood Project Meeting 2021

Maison du Lait, Paris

2nd July 2021

As a part of my doctoral studies, my research is motivated by the European ArtiSaneFood project. This was the first annual meeting among all the major French partners including Dr. Laurent Guillier (ANSES), Dr. Fanny Tenenhaus-Aziza (CNIEL), Prof. Emmanuel Vazquez (L2S), Prof. Julien Bect (L2S ) and representatives from cheese producing industrial partners.

• Find more on: ArtiSaneFood

Keywords: Quantitative Risk Assessment (QRA), Microbiology, Camembert de Normandy

Journée de doctorats

29 September 2021

This was the annual PhD students day (Journée de doctorats) at the Laboratoire de signaux et systemes (L2S), CentraleSupelec, Université Paris-Saclay. As a part of the doctoral studies all the first year students made a poster presentation. I presented my ongoing work on "Variance reduction with stratified sampling for bounded monotone function "

• Poster: click here

Keywords:Variance reduction, Monte Carlo, Stratified sampling

Presentation @ LOD 21

Grasmere, Lake District, England – UK

4-8 th October 2021

Our article "Numerical issues in maximum likelihood parameter estimation for Gaussian process interpolation ", with Sébastien Petit (Safran Aircraft Engines), Julien Bect and Emmanuel Vazquez (L2S, CentraleSupelec), was accepted at the 7th International Online and Onsite Conference on Machine Learning, Optimization, and Data Science - LOD 2021.

• Find more on: LOD 21

• Github repo: saferGPMLE

• Slides: click here

Keywords/Technologies Used: Gaussian Process, Machine learning, Optimization

AI4AI Hackathon

8 November - 15 December 2021

Me and my teammates, Fabien Girka, Manon Mottier and Lisa Poirier--Herbeck, reached the final of the #Hackathon #AI4IA (Artificial Intelligence For Industrial Application) challenge. Consequently, we had the opportunity to present our work in front of a jury of scientific experts and directions of SKF Group, ArianeGroup, Atos, Techniques de l'Ingénieur, MathWorks, Amazon Web Services (AWS), at Campus de l'Espace, in Vernon.

• Find more on: SKF Group

Keywords: Multivariate time series, Hackathon, AI4AI