Skip to content
SDGs Center UDINUS

Evidence record · UDINUS · SDGs

Modification of Reward Function in Reinforcement Q-learning Agent to Improve Quality of Service in Federated Edge Cloud

Rektorat UDINUS 2024 Published 2026-10-30

Evidence record · UDINUS · SDGs

Unit
Rektorat UDINUS
Period
2024
Goals
Record
50341d99-307d-4bde-94a9-479c7ad82a49
Published
2026-10-30

Universitas Dian Nuswantoro (UDINUS), through participating researchers identified in the source, conducted research on modifying the reward function used in Reinforcement Q-Learning in a Federated Edge Cloud (FEC) environment. The proposed modification used a logarithmic approach and a reduction variable. The study also introduced MIPS (Million Instructions per Second) and RAM (Random Access Memory) as additional variables in the reward function.

UDINUS evaluated several modifications to the reward function used in Reinforcement Q-Learning. The Quality of Services (QoS) evaluation used a workload dataset from PlanetLab. The evaluation found that the proposed reward function based on Logarithmic MIPS or RAM with a reduction parameter was able to improve SLA (Service Level Agreement) Violation by up to 10.6%. The approach was also followed by an improvement in SLA-PDM (Performance Degradation due to Migration) of up to 10%.

The source further reports that energy efficiency was preserved, consistent with the previous study referenced there, which had demonstrated efficient energy usage. Overall, the proposed reward-function modification had more influence on improving QoS performance, specifically SLA Violation and SLA-PDM, than on reducing energy use in the FEC environment. The activity is institutionally framed as UDINUS research on reward-function modification, QoS evaluation, and energy efficiency in a Federated Edge Cloud environment.

Reference https://www.scopus.com/pages/publications/85208116853?origin=resultslist