Masterarbeiten
Exploring Explainable Machine Learning and Enhancing Lower-Layer Triggered Mobility in Cellular Networks
Beschreibung
Goal: To obtain a rule-based heuristic from a well-performing AI agent for an NP-hard optimization problem.
AI models, especially deep learning systems, often operate as "black boxes," making complex decisions without revealing how they arrived at a specific conclusion. Explainable AI provides insights into the decision-making process, fostering transparency. As future 5G/6G networks evolve, mobile operators will increasingly rely on AI-driven solutions. For widespread adoption, it's crucial that operators trust these systems, a trust that is far more likely to be built if they can clearly understand how the AI arrives at its decisions.
First, we will focus on extracting the underlying rules that guide AI decision-making. Next, we will aim to develop a rule-based heuristic that can replicate the AI’s performance, offering a simpler yet equally effective solution.
Voraussetzungen
- Strong Python programming skills
- Interest in 5G/6G networks and mobility management
- Ability to work independently and learn new concepts