- Expertini Resume Scoring: Our Semantic Matching Algorithm evaluates your CV/Résumé before you apply for this job role: Stage Generative Privacy Attacks and Defenses in Federated Learning via Diffusion Models H/F.
Urgent! Stage - Generative Privacy Attacks and Defenses in Federated Learning via Diffusion Models H/F Job Opening In Gif-sur-Yvette – Now Hiring CEA
Description de l'offre
Context
Federated Learning (FL) enables multiple clients to collaboratively train machine learning models without sharing their data.
Instead, clients exchange local model updates with a central server, which uses them to improve a global model.
While this paradigm enhances data privacy, recent studies have shown that FL remains vulnerable to privacy breaches.
In particular, gradient inversion attacks can reconstruct sensitive client data from transmitted updates, posing serious privacy concerns [1].
Traditional approaches such as Deep Leakage from Gradients (DLG) [1], demonstrated that even simple models can leak identifiable information.
At the same time, diffusion models [3,4] have emerged as powerful generative frameworks capable of synthesizing realistic data from noisy or partial information.
Recent works demonstrate that diffusion models can enhance gradient-based privacy attacks [5] and inspire novel privacy-preserving strategies [6,7].
Objectives
The goal of this internship is to explore the dual role of diffusion models in attacking and defending Federated Learning systems:
[1] Zhu et al.
Deep Leakage from Gradients.
NeurIPS 2019.
[2] Zhao et al.
iDLG: Improved Deep Leakage from Gradients.
ICLR 2020.
[3] Ho et al.
Denoising Diffusion Probabilistic Models.
NeurIPS 2020.
[4] Song et al.
Score-Based Generative Modeling through SDEs. ICLR 2021.
[5] Gu et al.
Gradient-Guided Diffusion Models for Privacy Attacks.
2024.
[6] Liu et al.
DP-Fed-FinDiff: Differentially Private Federated Diffusion for Tabular Data.
2024.
[7] Chen et al.
Personalized Federated Diffusion with Privacy Guarantees.
2025.
[8] Fang et al.
GIFD: A Generative Gradient Inversion Method with Feature Domain Optimization ICCV 2023.
[9] Bonawitz et al.
Practical Secure Aggregation for Privacy-Preserving ML.
CCS 2017.
[10] Abadi et al.
Deep Learning with Differential Privacy.
CCS 2016.
Profil du candidat
Qu’attendons-nous de vous ?
The candidate should be in the last year of an engineering school or a master student (Bac+5) in a field related to machine learning/AI, who wishes to conduct research and development in an emerging, yet impactful field, in a collaborative environment.
The intern will work in a team of researchers, post-docs, and PhD students who are actively investigating various challenges and aspects of federated learning.
The candidate should have knowledge in machine learning and optimisation, and be skilled in Python programming and in using various machine learning libraries and frameworks.
Rejoignez-nous, venez développer vos compétences et en acquérir de nouvelles !
Vous avez encore un doute ?
Nous vous proposons :
Conformément aux engagements pris par le CEA en faveur de l'intégration des personnes handicapées, cet emploi est ouvert à toutes et à tous.
Le CEA propose des aménagements et/ou des possibilités d'organisation pour l'inclusion des travailleurs handicapés.
✨ Smart • Intelligent • Private • Secure
Practice for Any Interview Q&A (AI Enabled)
Predict interview Q&A (AI Supported)
Mock interview trainer (AI Supported)
Ace behavioral interviews (AI Powered)
Record interview questions (Confidential)
Master your interviews
Track your answers (Confidential)
Schedule your applications (Confidential)
Create perfect cover letters (AI Supported)
Analyze your resume (NLP Supported)
ATS compatibility check (AI Supported)
Optimize your applications (AI Supported)
O*NET Supported
O*NET Supported
O*NET Supported
O*NET Supported
O*NET Supported
European Union Recommended
Institution Recommended
Institution Recommended
Researcher Recommended
IT Savvy Recommended
Trades Recommended
O*NET Supported
Artist Recommended
Researchers Recommended
Create your account
Access your account
Create your professional profile
Preview your profile
Your saved opportunities
Reviews you've given
Companies you follow
Discover employers
O*NET Supported
Common questions answered
Help for job seekers
How matching works
Customized job suggestions
Fast application process
Manage alert settings
Understanding alerts
How we match resumes
Professional branding guide
Increase your visibility
Get verified status
Learn about our AI
How ATS ranks you
AI-powered matching
Join thousands of professionals who've advanced their careers with our platform
Unlock Your Stage Generative Potential: Insight & Career Growth Guide
Real-time Stage Generative Jobs Trends in Gif-sur-Yvette, France (Graphical Representation)
Explore profound insights with Expertini's real-time, in-depth analysis, showcased through the graph below. This graph displays the job market trends for Stage Generative in Gif-sur-Yvette, France using a bar chart to represent the number of jobs available and a trend line to illustrate the trend over time. Specifically, the graph shows 14340 jobs in France and 594 jobs in Gif-sur-Yvette. This comprehensive analysis highlights market share and opportunities for professionals in Stage Generative roles. These dynamic trends provide a better understanding of the job market landscape in these regions.
Great news! CEA is currently hiring and seeking a Stage Generative Privacy Attacks and Defenses in Federated Learning via Diffusion Models H/F to join their team. Feel free to download the job details.
Wait no longer! Are you also interested in exploring similar jobs? Search now: Stage Generative Privacy Attacks and Defenses in Federated Learning via Diffusion Models H/F Jobs Gif-sur-Yvette.
An organization's rules and standards set how people should be treated in the office and how different situations should be handled. The work culture at CEA adheres to the cultural norms as outlined by Expertini.
The fundamental ethical values are:The average salary range for a Stage Generative Privacy Attacks and Defenses in Federated Learning via Diffusion Models H/F Jobs France varies, but the pay scale is rated "Standard" in Gif-sur-Yvette. Salary levels may vary depending on your industry, experience, and skills. It's essential to research and negotiate effectively. We advise reading the full job specification before proceeding with the application to understand the salary package.
Key qualifications for Stage Generative Privacy Attacks and Defenses in Federated Learning via Diffusion Models H/F typically include Computer Occupations and a list of qualifications and expertise as mentioned in the job specification. Be sure to check the specific job listing for detailed requirements and qualifications.
To improve your chances of getting hired for Stage Generative Privacy Attacks and Defenses in Federated Learning via Diffusion Models H/F, consider enhancing your skills. Check your CV/Résumé Score with our free Resume Scoring Tool. We have an in-built Resume Scoring tool that gives you the matching score for each job based on your CV/Résumé once it is uploaded. This can help you align your CV/Résumé according to the job requirements and enhance your skills if needed.
Here are some tips to help you prepare for and ace your job interview:
Before the Interview:To prepare for your Stage Generative Privacy Attacks and Defenses in Federated Learning via Diffusion Models H/F interview at CEA, research the company, understand the job requirements, and practice common interview questions.
Highlight your leadership skills, achievements, and strategic thinking abilities. Be prepared to discuss your experience with HR, including your approach to meeting targets as a team player. Additionally, review the CEA's products or services and be prepared to discuss how you can contribute to their success.
By following these tips, you can increase your chances of making a positive impression and landing the job!
Setting up job alerts for Stage Generative Privacy Attacks and Defenses in Federated Learning via Diffusion Models H/F is easy with France Jobs Expertini. Simply visit our job alerts page here, enter your preferred job title and location, and choose how often you want to receive notifications. You'll get the latest job openings sent directly to your email for FREE!