Sparsity and interpretability?(Stanislav Böhm et al) (summarized by Rohin): If you want to visualize exactly what a neural network is doing, one approach is to visualize the entire computation graph of multiplies, additions, and nonlinearities. While this is extremely complex even on MNIST, we can make it much simpler by making the networks sparse, since any zero weights can be removed from the computation graph. Previous work has shown that we can remove well over 95% of weights from a model without degrading accuracy too much, so the authors do this to make the computation graph easier to understand.
Are models that are trained as sparse, rather than pruned to be sparse, different from each other? (Especially in terms of interpretability.)
This paper didn’t check that, but usually when you train sparse networks you get worse performance than if you train dense networks and then prune them to be sparse.
Are models that are trained as sparse, rather than pruned to be sparse, different from each other? (Especially in terms of interpretability.)
This paper didn’t check that, but usually when you train sparse networks you get worse performance than if you train dense networks and then prune them to be sparse.