4.1 Quantum Annealing for Machine Learning
Quantum boosting is the first QML algorithm we will consider in this book. This is also the algorithm that plays to the natural strengths of quantum annealing.
4.1.1 General principles of the QBoost algorithm
We start with the general principles of the Quantum Boosting (QBoost) algorithm before exploring a specific finance-related application. In the formulation of QBoost, we will be using the following definitions and notations:
| Object | Definition |
| xτ = (x1(τ),x2(τ),…,xN(τ)) | Vector of N variables (features) |
| yτ = ±1 | Binary label indicating whether xτ corresponds to Class 0 (y = −1) or Class 1 (y = +1) |
| {xτ,yτ}τ=1,…,M | Set of training events |
ci(xτ) = ± |
Value of the weak classifier i on the event τ |
| q := (q1,q2,…,qN) | Vector of binary (0 or 1) weights associated with each weak classifier |
Table 4.1: QBoost algorithm...
