“…Many of the quantum advantages associated with near-term variational QML algorithms relate to model capacity, expressivity and sample efficiency. In particular, variational QML algorithms may yield reductions in the number of required trainable parameters [ 214 ], generalization error [ 137 , 203 – 205 ], the number of examples required to learn a model [ 199 , 212 ] and improvements in training landscapes [ 137 , 199 , 203 , 207 , 208 , 252 ]. Evidence supporting one or more of these advantages has been found in both theoretical models and proof of principle implementations of quantum neural networks (QNNs) [ 137 , 203 , 204 , 207 ] and quantum kernel methods (QKMs) [ 199 , 201 , 205 ].…”