Abstract
Quantum routing plays a key role in the development of the next-generation network system. In particular, an entangled routing path can be constructed with the help of quantum entanglement and swapping among particles (e.g., photons) associated with nodes in the network. From another side of computing, machine learning has achieved numerous breakthrough successes in various application domains, including networking. Despite its advantages and capabilities, machine learning is not as much utilized in quantum networking as in other areas. To bridge this gap, in this article, we propose <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">a novel quantum routing model</i> for quantum networks that employs machine learning architectures to construct the routing path for the maximum number of demands (source–destination pairs) within a time window. Specifically, we present a deep reinforcement routing scheme that is called Deep Quantum Routing Agent (DQRA). In short, DQRA utilizes an empirically designed deep neural network that observes the current network states to accommodate the network’s demands, which are then connected by a qubit-preserved shortest path algorithm. The training process of DQRA is guided by a reward function that aims toward maximizing the number of accommodated requests in each routing window. Our experiment study shows that, on average, DQRA is able to maintain a rate of successfully routed requests at above 80% in a qubit-limited grid network and approximately 60% in extreme conditions, i.e., each node can be repeater exactly once in a window. Furthermore, we show that the model complexity and the computational time of DQRA are polynomial in terms of the sizes of the quantum networks.
Highlights
There are high demands of network resources and security in today’s network and the next-generation network systems since more and more devices are connected to the Internet and new services are created
We use two algorithms to train the deep network of Deep Quantum Routing Agent (DQRA), as a deep reward network, and as a deep Q network (DQN) [19]
We have modeled the problem of entanglement routing in quantum networks as a reinforcement learning problem that consists of input states, actions, and rewards
Summary
There are high demands of network resources and security in today’s network and the next-generation network systems since more and more devices are connected to the Internet and new services are created. There are currently no such works in the literature With such motivation, in this paper, we present a deep reinforcement routing scheme that is called Deep Quantum Routing Agent (DQRA). Our experiment study shows that, on average, both DQRA models are able to maintain a rate of successfully routed requests at 85–90% in a qubit-limited quantum network, and 60–75% in extreme conditions of network (i.e. all nodes can only be end nodes or repeater exactly once). We propose a reinforcement learning model for the problem with specific designs of environments, actions, and rewards, that guide the agents towards a schedule that fulfills the most traffic requests. 2) We present DQRA, a deep reinforcement routing scheme that consists of an empirically designed deep neural network that schedules requests and a qubitpreserved shortest path algorithm that routes selected ones.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
