Multi-Agent Reinforcement Learning for SLA-Aware Network Slicing in UAV-Enabled MEC

Collection of recent networking and ML-for-networks research: (1) UAV-enabled MEC: a predictive multi-agent RL (MAPPO) approach for SLA-aware trajectory control and compute allocation to reduce SLA violation probability/duration while saving energy. (2) Privacy-preserving intent fulfilment and assurance for 6G RAN/O-RAN: an architecture that provisions intents using only aggregate PM counters, proving an upper bound on traffic information leakage and defining intent-traffic unlinkability and node-opaque verification. (3) Transformer-based actor-critic for sequence-aware SFC partitioning to map