Using Machine Studying

Abstract: The Gamer's Non-public Network (GPN) is a consumer/server technology created by WTFast for making the community performance of online video games sooner and more dependable. GPN s use center-mile servers and proprietary algorithms to better connect online video-sport gamers to their recreation's servers throughout a large-space network. On-line video games are a massive leisure market and community latency is a key aspect of a player's competitive edge. This market means many different approaches to community architecture are carried out by completely different competing corporations and that those architectures are consistently evolving. Making Minecraft Server List between a consumer of WTFast and the online sport they wish to play is thus an extremely troublesome drawback to automate. Utilizing machine learning, we analyzed historical community information from GPN connections to explore the feasibility of community latency prediction which is a key part of optimization. Our next step will likely be to collect dwell information (together with consumer/server load, packet and port information and particular sport state information) from GPN Minecraft servers and bots. We will use this info in a Reinforcement Learning model together with predictions about latency to change the clients' and servers' configurations for optimum network performance. These investigations and experiments will improve the standard of service and reliability of GPN methods.