Q1: Explanation about the mismatch (1/4 and 1) between the theory (Theorem 2 and Corollary 2.1) and practice

We will answer the major points below and address all remaining ones in the final version. We leave further discussions on the convergence rate to future works. Eqn.11 in [B] is generic (a superset of most loss engineerings like [3, 29, A]), it uses bi-level We will add a discussion on [3, A, B] in the final version. Q2: Meta sampler has a similar idea to [12,24,27]. CIFAR10-L T); theirs are instance-based and ours is class-based (fewer parameters and simpler optimization landscape).