This page was translated by an AI (LLM) with a cursory human check and is awaiting full review.
ALDL: Deep Learning Launch Workshop
Manager : Kamel Guerda
Instructors : IDRIS engineers selected according to the chosen files.
The aim of this workshop is to help launch a project related to deep learning by offering a project review with an Artificial Intelligence engineer from IDRIS.
- Target audience
- Prerequisites
- Duration and practical info
- Programme
- Upcoming sessions
Target audience
This workshop is intended for projects:
- in a startup phase
- considering using deep learning as a tool to exploit their data
- without a deep learning engineer on their team
Prerequisites
You must have previously attended an introductory training to deep learning. Such a training is provided prior to the workshop: Practical Introduction to Deep Learning.
Ahead of the session, we request a file summarizing the project, the data, the objectives and perspectives as well as the questions that the team in charge of the project may have. This file allows the selection of projects as well as the preparation of the engineer assigned to the project for the session.
Files are evaluated in the order they are received.
Duration and practical info
This workshop lasts 1/2 day. Morning or afternoon depending on the availability of the engineers. The information will be communicated to you upon confirmation.
It takes place exclusively in-person at the premises ofIDRIS in Orsay (91).
Attendance
The maximum number of projects followed is 3. Multiple people can participate in the workshop around the same project.
Programme
During the session, discussions may cover various conceptual and technical aspects of the life cycle of a machine/deep learning project in order to:
- share the vision of a deep learning engineer on a specific scientific problem to understand the problems specific to deep learning
- transmit best practices related to data management (acquisition, preparation, analysis, ...)
- help in the understanding of certain specific models of the state of the art
- share best practices for training a model (language, library, hardware architecture)
- discuss the computing possibilities of Jean Zay
- determine the priority actions to take for an efficient start of the project
For smooth exchanges, it is necessary to bring your laptop in case you have documents or a demonstrator to show us.
To view the dates of the upcoming sessions for this workshop, go to the following page:
Registration
Registration is free but mandatory. To register, you must send the required file as a prerequisite by email to:
📥 kamel.guerda@idris.fr
or
📥 e-cours@idris.fr
For the presentation of your file, you can use the following template:
Project Presentation
No registration will be considered without prior submission of this file.