Files in this item

FilesDescriptionFormat

application/pdf

application/pdfSP20-ECE499-Thesis-Xie, Fengze.pdf (322kB)Restricted to U of Illinois
(no description provided)PDF

Description

Title:Multi-dimensional user scheduling by MDP and reinforcement learning
Author(s):Xie, Fengze
Contributor(s):Srikant, Rayadurgam
Subject(s):User Scheduling
Markov Decision Process
Reinforcement Learning
Abstract:In this project we try to simulate user scheduling in high dimension. We have two machines that can serve multiple users at one time. Each machine has two-dimensional spaces, the memory and the CPU. There are two types of user that take different amounts of resource. Each user will arrive at the system and stay in the machine in a Poisson process. We want to maximize the average number of users in the system by using MDP and reinforcement learning separately.
Issue Date:2020-05
Genre:Other
Type:Other
Language:English
URI:http://hdl.handle.net/2142/107236
Date Available in IDEALS:2020-06-11


This item appears in the following Collection(s)

Item Statistics