Abstract of the talk// Aqueous solutions form the molecular backdrop of countless physical, chemical, and biological processes. Yet despite decades of study, the microscopic fluctuations that govern their thermodynamic and dynamical behavior remain only partially understood. Much of our conceptual framework for water relies on low-dimensional intuition—hydrogen-bond networks, structural motifs, and simplified species pictures—that may obscure the true complexity of its fluctuations.

In this talk, Ali will present recent efforts from his group to move beyond intuition-driven models toward a data-driven framework for interrogating liquid water. He will introduce an unsupervised learning protocol designed to quantify and interpret high-dimensional structural and dynamical fluctuations without imposing predefined collective variables.

Using this approach, he will revisit several canonical problems: the proposed coexistence of high- and low-density liquid motifs at ambient and supercooled conditions; the structure of the hydrated excess proton beyond the limiting Eigen and Zundel pictures; and, more generally, how ions perturb water structure as so-called makers or breakers. In each case, Ali will assess both the insights and the limitations of traditional low-dimensional descriptions.

The talk will conclude with a discussion of how integrating data science with chemical physics enables us not merely to analyze fluctuations, but to refine and reconstruct our foundational concepts of liquids and collective behavior.