New answers tagged

1

First of all, for a lot of realistic problems, the fitness function evaluation is usually orders of magnitude greater in complexity than the rest of the genetic algorithm. This is not always true, but often is true (e.g. imagine trying to optimise a simulation where you need to execute the simulation completely to obtain the fitness). So optimising the GA ...


0

When input is fed into a neural network, the neural network essentially is doing a large number of matrix multiplications. If there are many many small non-zero weights, there are a lot of multiplications to do. If those weights are rounded down to zero, then obviously the multiplications don't need to happen so compute time is saved. Another benefit is that ...


Top 50 recent answers are included