3eb2f1a06667bfb9daba7f7effa0284b-AuthorFeedback.pdf

We thank the reviewers for their careful consideration and their feedback, our short replies are provided below. General response for Reviewers 1 & 2. -"Can the results be extended to convex loss functions?" . Reviewer 1. -"Could you remove the assumption that variance of quantization σ We will add this discussion in the revised paper. Size of the neural network is too small. If the ReLU activation was used, the tested loss function is not smooth.