Kvantové strojové učenie: hudba budúcnosti, alebo len rozmarný nápad?
Quantum machine learning: is this the future or is this just fantasy?
The promise of quantum machine learning is causing a lot of excitement in academia and industry alike. Despite bold claims, there are still significant challenges to overcome before quantum machine learning could beat its classical cousin. In this talk, I will focus on two common types of quantum neural network architectures and explain some of the problems in their trainability. Notably, many of the proposed quantum machine learning models exhibit a barren plateau phenomenon that wipes out any potential speedups. I will conclude by proposing a training method that avoids the existing no-go results.