Simulation of DRL (Deep Re-inforcement Learning)

In Software defined networks, re-routing under predefined conditions by monitoring the status of each data flow can be done through dynamic routing. But the problems of monitoring duration and lack of learning ability from its previous experiences exists. Hence the aim of this project is to implement a reinforcement based learning agent(Which can learn from its previous experience) to perform efficient routing over the network and hence improve the traffic management in SDN's