5a29503a4909fcade36b1823e7cebcf5-AuthorFeedback.pdf

We would like to thank all referees for their appreciation of our results and the useful feedback. Reviewer 1: Thank you for pointing us to the relevant references. Si et al. was made available only in July 2020 during ICML, while we cited Hu et al. as Reference [17] in our paper. Section 5.3 of Hu et al. focus solely on loss functions which are linear in Y. On the contrary, our loss function is linear The results from Hu et al. thus are not directly applicable to Our proof is also novel: it invokes duality results on the expectation parameter space. To our best knowledge, Faury et al. and Si et al. provide convergence results for the objective value, not for the solution.