While machine learning models are typically trained to solve prediction
problems, we might often want to use them for optimization problems. For
example, given a dataset of proteins and their corresponding fluorescence
levels, we might want to optimize for a new protein with the highes