News

IEEE8h
Dynamic Semantic Compression for CNN Inference in Multi-Access Edge Computing: A Graph Reinforcement Learning-Based Autoencoder - IEEE Xplore
This paper studies the computational offloading of CNN inference in dynamic multi-access edge computing (MEC) networks. To address the uncertainties in communication time and edge servers’ available ...
IEEE18h
Physical Layer-Aware Route and Spectrum Allocation in Optical Networks by Multi-Objective Deep Reinforcement Learning - IEEE Xplore
We propose a multi-objective Reinforcement Learning approach for RSA in optical networks, integrating awareness of signal quality, delay, and load-balancing. It outperforms traditional algorithms with ...