The paper proposes a new MIP formulation for setting reserve prices in in auctions. While the focus of the work is on the edges of the interest of NeurIPS, the paper is well-written and the algorithmic approach is novel and solid. The new formulation results in good experimental results. We urge the authors to incorporate clarifications on the importance of their work the the ML community into the main paper, e.g. into the abstract and the introduction. The paper would also benefit from a discussion of how the proposed algorithm can be included in practical scenarios.