… The authors would like to thank the anonymous referees of STOC 2020 for their helpful
comments and suggestions. The authors would like to express their sincere gratitude to matrix …

… As of the time of preparing this camera ready version (April 13, 2020), we are not aware of
any fix to that … We are also grateful to the reviewers of STOC 2020 for their helpful comments. …

… State-of-the-art results on image recognition tasks are achieved using over-parameterized
learning algorithms that (nearly) perfectly fit the training set and are known to fit well even …

… To conclude, let us comment on how we prove the reduction step (Lemma 3.6). The main
idea is to use an encoding lemma, inspired by Razborov’s proof of Håstad’s switching lemma […

We study differentially private (DP) algorithms for stochastic convex optimization: the problem
of minimizing the population loss given iid samples from a distribution over convex loss …

… Still, the (1/2 + 1/polylog(n))-inapproximability result is not enough to get us a non-trivial (say…
2020. New lower bounds for probabilistic degree and AC0 with parity gates. Electronic …

… Compared to the results for (k,z)-clustering in [Feldman and Langberg, STOC 2011], our
framework saves a ε 2 d factor in the coreset size. Compared to the results for (k,z)-clustering in […

… Beyond allowing us to address basic questions about which error rates are achievable in
polyno… For now, let us note that our subroutine solves Problem 1.9 when Σ𝑖 is the empirical …

… al., STOC ‘15]. We also extend our techniques to solving the problem on vertex-weighted
graphs … Before explaining how we circumvent this issue, let us define a maximal tight chain. …

… To understand the motivation behind our main idea, let us first consider a slightly modified
version of the classic ES-trees that achieves the same running time: We maintain for each …