Investigating Parameter-Efficiency of Hybrid QuGANs Based on Geometric Properties of Generated Sea Route Graphs
This work addresses the need for parameter-efficient generative models in quantum computing applications, such as generating shipping data, but it is incremental as it builds on existing QuGAN methods with a specific domain focus.
The study tackled the problem of generating artificial shipping route graphs using hybrid quantum-classical generative adversarial networks (QuGANs), finding that QuGANs can learn geometric properties quickly but struggle with variance, and some achieve similar quality to larger classical GANs with fewer parameters.
The demand for artificially generated data for the development, training and testing of new algorithms is omnipresent. Quantum computing (QC), does offer the hope that its inherent probabilistic functionality can be utilised in this field of generative artificial intelligence. In this study, we use quantum-classical hybrid generative adversarial networks (QuGANs) to artificially generate graphs of shipping routes. We create a training dataset based on real shipping data and investigate to what extent QuGANs are able to learn and reproduce inherent distributions and geometric features of this data. We compare hybrid QuGANs with classical Generative Adversarial Networks (GANs), with a special focus on their parameter efficiency. Our results indicate that QuGANs are indeed able to quickly learn and represent underlying geometric properties and distributions, although they seem to have difficulties in introducing variance into the sampled data. Compared to classical GANs of greater size, measured in the number of parameters used, some QuGANs show similar result quality. Our reference to concrete use cases, such as the generation of shipping data, provides an illustrative example and demonstrate the potential and diversity in which QC can be used.