Join our Team
About this opportunity
We’re looking for an enthusiastic MSc student to join our team and contribute to experimenting advanced machine learning for telecom scheduler performance;
In this thesis project, you will have all the freedom to explore different advanced machine learning technique for identified scheduler use case (i.e. link adaption), such as continual learning with Elastic Weight Consolidation for Evolving Radio Conditions , as well as apply Multi-objective Reinforcement learning for joint control: formulate joint MCS, rank, and power control as multi-objective RL balancing throughput, BLER, and energy, report Pareto fronts, and study robustness across interference and mobility regimes.
You will gain hands-on experience working with real industrial data and contribute to methods with direct business value.
What you will do
- Conduct a literature review to learn deep on many advance machine learning techniques, firm opinion of applicable optimizations tailored to our use cases
- Design, implement, and evaluate the RAN performance with advanced deep learning methodologies.
- Benchmark AI model performance against our currently implemented methods
- Build an end-to-end prototype for deployment in an industrial setting
- Summarize your work in a written thesis and present your results to the team
The skills you bring:
We are looking for a master’s student who:
- Very curious about new technology, Quick learner with open mind
- Is currently enrolled in a quantitative MSc program such as Computer Science, Data Science, Communication, or a related field.
- Has hands-on experience with Python programming and common ML libraries (NumPy, Pandas, scikit-learn, etc.)
- Ideally has practical experience with deep learning frameworks (e.g., JAX, TensorFlow, PyTorch)
- Ideally has experience or strong interests with Random access network
- Communicates well in English and enjoys working in a collaborative environment.
Read Full Description