If the problem is that the class probabilities change too rapidly wrt the input data, then maybe a solution would be to add a term penalizing large gradients to

If the problem is that the class probabilities change too rapidly wrt the input data, then maybe a solution would be to add a term penalizing large gradients to the cost used for backpropagation.