Optimizing Kernel-Target Alignment for cloud detection in multispectral satellite images
This work addresses resource efficiency in quantum machine learning for cloud detection in remote sensing, but it is incremental as it builds on existing Kernel-Target Alignment methods with a toy model study.
The authors tackled the problem of optimizing Kernel-Target Alignment to reduce hardware resources in quantum classifiers by studying its landscape with a toy model, finding that underparameterized circuits have many local extrema or flat landscapes with narrow global extrema, and they determined the peak width depends on data amount, using multispectral satellite data for cloud detection.
The optimization of Kernel-Target Alignment (TA) has been recently proposed as a way to reduce the number of hardware resources in quantum classifiers. It allows to exchange highly expressive and costly circuits to moderate size, task oriented ones. In this work we propose a simple toy model to study the optimization landscape of the Kernel-Target Alignment. We find that for underparameterized circuits the optimization landscape possess either many local extrema or becomes flat with narrow global extremum. We find the dependence of the width of the global extremum peak on the amount of data introduced to the model. The experimental study was performed using multispectral satellite data, and we targeted the cloud detection task, being one of the most fundamental and important image analysis tasks in remote sensing.