Optimization over Trained (and Sparse) Neural Networks: A Surrogate within a Surrogate

ArXi:2505.01985v3 Announce Type: replace-cross In constraint learning, we use a neural network as a surrogate for part of the constraints or of the objective function of an optimization model. However, the tractability of the resulting model is heavily influenced by the size of the neural network used as a surrogate. One way to obtain a tractable surrogate is by pruning the neural network first.