Total Positivity

Abstract.
A totally positive matrix is a real matrix all of whose minors are positive. My lectures will give a broad survey of the relation between total positivity and combinatorics:
  1. Total positivity and networks: I will discuss the parametrization of totally positive matrices dating back to Loewner and Whitney, and the work of Brenti and Lindström-Gessel-Viennot relating total positivity to non-intersecting paths in networks. I will then discuss the generalization of total positivity to Grassmannians in the works of Lusztig and Postnikov.
  2. Total positivity and statistical mechanics: there is a close relation between total positivity and certain statistical mechanical models (dimer model, electrical networks, Ising model) on planar graphs. I will survey some of these connections, following works of Postnikov, Curtis-Ingerman-Morrow, de Verdier-Gitler-Vertigan, Kenyon-Wilson, Lam, Lis, Galashin-Pylyavskyy.
  3. Total positivity and combinatorial topology: in recent years an analogy between totally positive spaces and the theory of convex polytopes has been appearing. I will talk about the motivations from poset topology (e.g. work of Björner, Fomin-Shapiro, Hersh) and scattering amplitudes (Arkani-Hamed and Trnka), and some new results of Galashin, Karp, and myself in this direction.