search

aryadas98 / cs359-networking-project

Applying machine learning for tcp congestion control
MIT License
4 stars 2 forks source link

RL instead of Polynomial Regression #1

Open Zeee-D opened 2 years ago

Zeee-D commented 2 years ago

Hi, you are doing great work. Can we use reinforcement learning instead of a regression model to find the congestion window in an adaptive manner?

aryadas98 commented 2 years ago

Hi, yes, reinforcement learning would be the ideal way to solve the problem. It would attempt to find an optimal solution. We weren't proficient in RL at that time, so we did not try it. All the best if you solve it! ᐧ

On Tue, Jan 25, 2022 at 6:25 PM Zeee-D @.***> wrote:

Hi, you are doing great work. Can we use reinforcement learning instead of a regression model to find the congestion window in an adaptive manner?

— Reply to this email directly, view it on GitHub https://github.com/aryadas98/cs359-networking-project/issues/1, or unsubscribe https://github.com/notifications/unsubscribe-auth/ACOC4LIHTHR7HGC6BSWRDODUX2MU3ANCNFSM5MYFWGVQ . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you are subscribed to this thread.Message ID: @.***>

Zeee-D commented 2 years ago

Hi arya, I have one doubt regarding your code. Here you have defined the buffer size as 16. But in the csv file you have shown window size more than that, up to 450. How is that possible ? Also what is the packet size?