Scalable and Cost Efficient Maximum Concurrent Flow over IoT using Reinforcement Learning.
Abou-Bakr DjakerKechar BouabdellahHatem Ibn-KhedherHassine MounglaHossam AfifiPublished in: IWCMC (2020)
Keyphrases
- cost efficient
- reinforcement learning
- hierarchical reinforcement learning
- management system
- machine learning
- web scale
- governmental organizations
- supervised learning
- learning algorithm
- function approximation
- temporal difference
- state space
- flow patterns
- neural network
- optimal policy
- dynamic programming
- highly scalable
- concurrent execution
- policy search
- multi agent
- temporal difference learning
- reinforcement learning algorithms
- reward function
- markov decision processes
- lightweight