Abstract
In this thesis, a Variational Quantum Boltzmann Machine(VarQBM) employing the use of Variational Quantum Imaginary Time Evolution(VarQITE) to prepare approximated Gibbs states was studied. The VarQBM in addition to classical Restricted Boltzmann machines(RBMs) were used to find underlying trends in training data. A method of encoding feature values into a VarQBM is proposed utilizing a neural network feature engineering scheme, compressing multiple features into relatively few qubits of choice, which was chosen to be named NN-VarQBM.