Consider that we want to create a very big neural network. If we consider to use dense layers, we might face some challenges. Now consider that we use sparse layers instead of dense layers. When using a really sparse model, we would have much less parameters and could create much bigger neural networks. Is there an efficient algorithm to parallelize the training of such networks?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.