4.2 QBoost Applications in Finance
Quantum Annealing for Machine Learning (QAML) has been applied productively to a wide range of financial and non-financial use cases. It demonstrated a performance advantage in comparison with standard classical machine learning models such as the binary decision tree-based Extreme Gradient Boosting (XGBoost) and Deep Neural Network (DNN) classifiers, especially on relatively small datasets. The QAML use cases come from such diverse fields as high-energy physics (the Higgs boson detection [233]) and computational biology (the classification and ranking of transcription factor binding [200]). In finance, the most obvious application of QAML is to credit scoring and fraud detection as well as to the construction of strong trading signals from large numbers of weak binary (buy/sell) trading signals.
In this section, we analyse QBoost performance on the more conventional binary classification problem – forecasting credit card client defaults...
