Why am I excited about math formalization with AI assistance? Paradoxically, it might free up human time and exposition space for some more human, softer and often neglected aspects of mathematics - intuition, storytelling, experimentation, etc. (From arxiv.org/pdf/2603.26524)

🥤"Optimizing mixtures of metal–organic frameworks for robust and bespoke passive atmospheric water harvesting" by C. Harriman, Q. Ke, T. Vlugt, A. Howarth, C. Simon. feedback welcome on our new preprint! 💡the key idea is to optimize the composition of *mixtures* of metal-organic frameworks (MOFs) to comprise the adsorbent bed of a passive atmospheric water harvester. with a mixture, each MOF can specialize in delivering water under different conditions. we frame a linear program for the optimal design of the composition of the MOF adsorbent bed that is: (1) bespoke, i.e., tailored to a declared geographic region and time frame and (2) robust, i.e. satisfies minimum daily water delivery requirements despite variance in weather. both weather data and MOF water adsorption data are inputs to our optimization problem.

Among the most common questions that I get from data scientists is "Are there any compelling applications of Topological Data Analysis?" The answer is yes, with high multiplicity. Would you like to see a brand new one? 1/