Support system for designing an artificial intelligence application, executable on distributed computing platforms
AssigneeBull SAS
InventorsFrançois EXERTIER, Mathis GAVILLON
Life of this patent
Estimating the term…
Abstract
The present invention relates to a learning engine function and the use thereof in a system using a suite of modular and clearly structured Artificial Intelligence application design tools (SOACAIA), executable on distributed or undistributed computing platforms to browse, develop, make available and manage AI applications, this set of tools implementing four functions: A Studio function; A Forge function; An Orchestrator function; A fast machine learning engine FMLE (FastML Engine) function.
Claims28
A system using a suite of modular and clearly structured Artificial Intelligence application design tools (SOACAIA), executable on computing platforms or hybrid infrastructures to browse, develop, make available and manage Artificial Intelligence (AI) applications, this set of tools implementing four functions:
- a Studio function making it possible to establish a secure and private shared space for the company wherein the extended team of business analysts, data scientists, application architects and IT (Information Technology) managers can communicate and work together collaboratively;
- a Forge function making it possible to industrialize AI instances and make analytical models and their associated datasets available, via a catalog, to the development teams, subject to compliance with security and processing conformity conditions;
- an Orchestrator function for managing the total implementation of the AI instances designed by the STUDIO function and industrialized by the Forge function and to perform permanent management on a hybrid cloud infrastructure;
- a fast machine learning engine (FMLE) function, which is a machine learning toolbox for high-performance computers HPC and hybrid infrastructures, facilitating the use of computing power and the possibilities of high-performance computing clusters as execution support for machine learning training models and specifically deep learning training models.
(canceled)
(canceled)
(canceled)
(canceled)
(canceled)
A fast machine learning engine (FastMLE) function comprising an interface, the aim of which is, during its execution on a platform, to facilitate the phase of designing and fine-tuning machine learning models when a first draft of this model is produced and to train on one or more annotated dataset(s) in order to fine-tune the model, the FastMLE module making it possible to manage the existing annotated and ready-to-use dataset files, by enabling the selection of one or more dataset files by the “data scientist” and loading them onto an infrastructure where the training of their model is to be executed, or to save them in at least one database operated by FastMLE.
Dependencies are inferred from the claim text.