Experimental Study of Context Based Routing Using Deep Reinforcement Learning
Rutgers, the State University of New Jersey
Through the proposed experimental campaign we intend to instigate research on the following issues: Can artificial intelligence (AI)—and in particular deep reinforcement learning (RL)—with the help of contextual information achieve real-time or near-real time packet routing and under which conditions? Can deep RL achieve that in a distributed manner? Is deep RL sufficiently resilient and adaptive in solving the super complex NGI routing problems?