PhD “Convergence of Graph and Vector Approaches for Network Digital Twin Simulation”

Job title:

PhD “Convergence of Graph and Vector Approaches for Network Digital Twin Simulation”

Company:

Orange

Job description

about the roleContext: Your role will be to conduct a PhD research project on: « the convergence of graph and vector-based approaches to optimize simulation tools for network digital twins ».This is part of the Orange Thing’in digital twin management platform, aiming to improve the management of Orange’s networks. The automation of these complex networks can be enhanced by learning algorithms that detect and correct errors [1].Recent advances in machine learning on graphs and the emergence of Graph Neural Networks (GNNs) open new perspectives for analyzing networks [2]. Recent models can also consider the dynamics of graphs [3]. These models capture both local and global information, which is useful for understanding complex networks. In the context of Orange’s networks, these techniques will enable the prediction of customer traffic, optimize resource allocation, and simulate scenarios such as failures or traffic spikes.Problematic: Despite efforts to optimize the interaction between relational databases and machine learning APIs [4], graph databases are less studied. Graph learning pipelines often use external libraries but lack centralized governance and the exploitation of past experiences. Models are often retrained from scratch, increasing costs. Integrating learning into graph databases would better track data evolution.Scientific Objective – Results and Challenges to Overcome: The goal of the PhD thesis is to develop a hybrid database combining graphs and vectors to optimize prediction and simulation of telecom networks, while improving graph queries with machine learning. The expected prototype will natively integrate machine learning functionalities through a hybrid graph and vector representation.Expected Results and Challenges to Overcome:

• Multi-modeling: Representing data as graphs and vectors. Challenges: Ensuring consistency between these representations.
• Hybrid Querying: Defining a language that effectively handles graphs, vectors, and learned data. Challenges: Optimizing queries for these different representations.
• Storage and Historical Data: Implementing coherent storage for data, models, and vectors. Challenges: Managing dependencies, historical data and models, and reducing storage space while maintaining good performance.

References:
[1] Ngo, D.-T., et al. (2023). Empowering Digital Twin for Future Networks with Graph Neural Networks. Future Internet, 15(12), 377.
[2] Jiang, W. (2022). Graph-Based Deep Learning for Communication Networks. Computer Communications, 185, 40-54.
[3] Zheng, Y., Yi, L., & Wei, Z. (2025). A Survey of Dynamic Graph Neural Networks. Frontiers of Computer Science, 19, 196323.
[4] Kläbe, S., et al. (2023). In-Database ML Approaches. EDBT, 23.about youRequired Scientific and Technical Skills, and Personal QualitiesThe required skills correspond to those expected at the Master’s or Engineering school level in Computer Science. An experience in one or more of the following areas will be highly valued:Artificial Intelligence: Machine Learning, (Graph) Neural Networks, embeddings
Real-time analytics systems and frameworks, and/or graph processing (Pregel, NetworkX, Spark)
NoSQL databases, Graph databases
Distributed systems, Big Data
Semantic Web
For the implementation of prototypes, algorithm design, and validation through simulations, strong programming skills are essential.The candidate should demonstrate curiosity and autonomy. Proficiency in English (written and spoken) is required, while French is a plus but not mandatory.Required EducationMaster’s degree or Engineering diploma in Computer Science
Desired ExperienceProjects and/or internships related to the skills mentioned above will be highly appreciated.additional information· You will join a multidisciplinary team and benefit from the expertise of its developers, researchers, and project managers.· You will work on a collaborative research project that fosters partnerships with various industrial and academic partners, providing an opportunity to develop your analytical skills and creativity.As digital twin technologies are central to the development of the Thing’in the Future platform, your work could be highlighted by Orange as an innovative and differentiating element in its future services.departmentThe ambition of the Innovation Division is to take Orange’s innovation even further and strengthen its technological leadership, by mobilising our research capabilities to nurture responsible innovation that serves people, informs the Group’s long-term strategic choices and influences the global digital ecosystem.
We train the experts in today’s and tomorrow’s technologies, and ensure continuous improvement in the performance of our services and our efficiency. The Innovation division employs 6,000 people worldwide dedicated to research and innovation, including 740 researchers. With a global vision and a wide range of profiles (researchers, engineers, designers, developers, data scientists, sociologists, graphic designers, marketers, cybersecurity experts, etc.), the men and women of Innovation listen to and serve the countries, regions and business units to make Orange a trusted multiservice operator.Within Innovation, you will be part of a research team at the forefront of innovation and expertise in the fields of ‘Ambient Computing’ and ‘Digital Twins’. More specifically, you will be working with members of the team developing the Thing’in the future research platform. This platform offers key functions for building digital twins.contractThesis

Expected salary

Location

Cesson-Sévigné, Ille-et-Vilaine

Job date

Thu, 03 Apr 2025 22:46:18 GMT

To help us track our recruitment effort, please indicate in your email/cover letter where (vacanciesin.eu) you saw this job posting.