Work

LLM-Based Algorithm Selection For Network Optimization and Orchestration

Year: 2025

Type: article

Abstract: A modular LLM-based framework that uses contextual reasoning to dynamically select the most suitable optimization algorithm for real-time network orchestration. It addresses scaling challenges, reduces manual decision-making, and improves SLA-aware service partitioning across distributed infrastructures. Presented live at the SC25 Conference INDIS Workshop.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.17801570

Author: Dalgkitsis, Anestis

Institution: University of Amsterdam

Language: English

Topic: Software-Defined Networks and 5G

Subfield: Computer Networks and Communications

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: European Commission

Award: Deep Programmability and Secure Distributed Intelligence for Real-Time End-to-End 6G Networks