"In machine learning, the goal is not just to fit the data, but to find a model that captures the underlying structure of the data."
"The only way to discover the limits of the possible is to go beyond them into the impossible."