Training quantum machine learning models on cloud without uploading the data
This work addresses data security concerns for users of quantum cloud services, offering a novel approach to privacy-preserving quantum machine learning.
The paper tackles the problem of data privacy in quantum machine learning by enabling dataset owners to train models on cloud quantum platforms without uploading their data, achieving a reduction in required circuit depth from O(2^n) to O(n) and allowing trained models to run on classical computers.
Based on the linearity of quantum unitary operations, we propose a method that runs the parameterized quantum circuits before encoding the input data. This enables a dataset owner to train machine learning models on quantum cloud computation platforms, without the risk of leaking the information about the data. It is also capable of encoding a vast amount of data effectively at a later time using classical computations, thus saving runtime on quantum computation devices. The trained quantum machine learning models can be run completely on classical computers, meaning the dataset owner does not need to have any quantum hardware, nor even quantum simulators. Moreover, our method mitigates the encoding bottleneck by reducing the required circuit depth from $O(2^{n})$ to $O(n)$, and relax the tolerance on the precision of the quantum gates for the encoding. These results demonstrate yet another advantage of quantum and quantum-inspired machine learning models over existing classical neural networks, and broaden the approaches to data security.