Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
April 1, 2026IET conference proceedings.

Automatic regulation and performance optimization method of computer network experimental environment based on deep reinforcement learning

View Full Paper
Ask AI
Bookmark
Share

Authors

CLC Chunliang Li

Discussion

Loading...

Member takes

Overview

Demonstrates an automatic regulation method for optimizing network performance, indicating significant improvements in throughput and delay.

Key Points

  • The central aim is to address the regulation challenges in dynamic computer network experimental environments using deep reinforcement learning.
  • Developed a topology aware GNN-MAPPO framework integrating graph neural networks and multi-agent proximal policy optimization.
  • Designed a dynamic weighted reward mechanism to balance multi-objective network performance metrics.
  • Conducted experiments on the OMNeT++ platform employing the Fat-Tree topology.
  • Improved throughput by 39% compared to traditional methods.
  • Reduced delay by 38% relative to existing frameworks.
  • Decreased packet loss rate by 66% when compared to conventional techniques.

Cite This Study

C Chunliang Li (2026) studied this question.

synapsesocial.com/papers/69ccb72e16edfba7beb8901ahttps://doi.org/10.1049/icp.2026.0253
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1A deep reinforcement learning approach for flexible multipath routing using graph neural networks2025
  2. 2A deep reinforcement learning approach for flexible multipath routing using graph neural networks2025 · 1 citations
  3. 3Integrating Reinforcement Learning and LLM with Self-Optimization Network System2025 · 1 citations
  4. 4Deep reinforcement learning for SDN routing optimization enhanced by graph multi-head attention2026
  5. 5Hybrid Graph Neural Network–Reinforcement Learning Framework for Intelligent Programmable Network Automation2026