Internship - High Precision Interpretable Machine

Position description

**Category**:

- Mathematics, information, scientific, software

**Contract**:

- Internship

**Job title**:

- INTERNSHIP - High Precision Interpretable Machine Learning - 6 months - Saclay H/F

**Subject**:

- Interpretability and High Precision Training for Neural Networks

**Contract duration (months)**:

- 6

**Job description**:

- At the_ Institute of Applied Sciences and Simulation for Low-Carbon Energies_ (ISAS) of the CEA, we focus on research and innovation in **analytical sciences**. As data analysis plays a pivotal role, we are interested in methodological advancements in **statistics**, **mathematics** and **computer science**, for instance, via the development of state-of-the-art AI models, adapted to our needs.-
- The internship targets the exploration of the **state-of-the-art** and the development of **optimisation techniques** for neural networks. The objective is to find possible ways to increase precision and interpretability of deep learning algorithms. In particular, we shall focus on the following tasks:

- ** critical review** the state-of-the-art in neural network optimisation to better understand the critical aspects playing a role in neural network precision;
- analysis of **second order** neural network optimisers for reliable and interpretable machine learning in physics;
- generalisation of some proposed techniques to enhance precision in neural network predictions.- The internship will be a collaboration between the DES (_Direction of Energies_) and the DRF (_Direction of Fundamental Research_) of CEA. The intern will be hosted by the _Laboratory of Artificial Intelligence and Data Science_ (LIAD) at the DES, in collaboration with the Institute of Theoretical Physics (IPHT).**Methods / Means**:

- optimisation, deep learning, machine learning, AI, physics

**Applicant Profile**:

- We look for a passionate student at the end of their studies (e.g. the French M2 level), with a good understanding of **machine learning** and **coding techniques**. Good knowledge of any **deep learning framework** (PyTorch, JAX, Tensorflow) in Python is mandatory, as well as abiding to good object
- oriented coding practices. A basic understanding of physics (statistical mechanics) is appreciated and considered a plus, though not necessary.Position location

**Site**:

- Saclay

**Job location**:

- France, Ile-de-France, Essonne (91)

**Location**:

- Saclay

**Languages**:

- French (Fluent)
- English (Intermediate)

Requester

**Position start date**:

- 01/01/2025

General information

**Organisation**:
The French Alternative Energies and Atomic Energy Commission (CEA) is a key player in research, development and innovation in four main areas:

- defence and security,
- nuclear energy (fission and fusion),
- technological research for industry,
- fundamental research in the physical sciences and life sciences.

Drawing on its widely acknowledged expertise, and thanks to its 16000 technicians, engineers, researchers and staff, the CEA actively participates in collaborative projects with a large number of academic and industrial partners.

The CEA is established in ten centers spread throughout France

**Reference **:2024-34335**Description de l'unité**:

- Notre Service dédié au Génie Logiciel pour la Simulation (SGLS) réalise et maintient des plateformes génériques, pérennes et open source dans le but:

- d'exploiter les codes de calculs à l'aide d'outils de mise en données, prétraitements et postraitements, standards ou spécifiques;
- de fournir aux physiciens les méthodes et outils leur permettant d'optimiser leurs conceptions et de traiter les incertitudes de leurs études de sureté.

Le Laboratoire d'Intelligence Artificielle et de science des Données (autrement nommé le LIAD) réalise et maintient une plateforme générique, pérenne et open source pour fournir à nos physiciens des méthodes et outils leur permettant d'améliorer leurs modèles, d'optimiser leurs conceptions et de traiter les incertitudes de leurs études : la plateforme Uranie.

Uranie ? Oui, notre plateforme permet dans l'approche VVQI (Validation, Vérification et Quantification d'Incertitude) de créer des plans d'expériences adaptés aux besoins d'une analyse de sensibilité, d'un problème d'optimisation ou de la génération d'une base d'apprentissage ou de test pour un modèle de substitution.

Uranie permet de piloter le lancement des codes ou fonctions de manière séquentielle ou avec différentes approches de parallélisation.