
School of Mathematics
Northwest University, Xi'an, Shaanxi Province, China
Time Zone Notice. All scheduled times are in U.S. Central Standard Time (CST, UTC−6). Each invited presentation receives 30 minutes (25-minute talk plus 5-minute discussion); the keynote receives 60 minutes. All seven scheduled presentations take place on Saturday, November 7, 2026, from 8:00 AM–12:00 PM CST.
Krein-Feller operators arise naturally in analysis on fractals. It has been proved that under suitable conditions on a measure, there exists an orthonormal basis of the L^2 space of the measure consisting of eigenfunctions of the operator. We study the continuity of these eigenfunctions. The problem is related to the nodal sets of the eigenfunctions. This is a joint work with Wen-Quan Zhao.
We generalize the concept of cyclic contraction and present some results about related iterated function systems
A self-similar iterated function system (IFS) on \(\mathbb{R}\) is a finite family \(\mathcal{F}=\{f_i\}_{i=1}^{m}\) of contracting similarities \(f_i:\mathbb{R}\to\mathbb{R}\). Its attractor is the unique nonempty compact set \(\Lambda\) satisfying \(\Lambda=\bigcup_{i=1}^{m}f_i(\Lambda)\). We consider the overlapping case, in which the sets \(f_i(\Lambda)\) are not pairwise disjoint. This talk focuses on results from two recent joint papers and the separation conditions underlying these results. First, in joint work with B. Bárány and A. Śpiewak, we investigate the complete multifractal formalism for self-similar measures in the overlapping setting. Second, in joint work with J. C. Xiao, we investigate the existence of points in the attractor with infinitely many codings when the similarity dimension is less than one. Finally, we discuss new separation conditions that have proved useful in these results and are applicable in more general settings.
Expanding Thurston maps are branched covering maps of the 2-sphere with expanding dynamics. They include all postcritically finite rational maps whose Julia set is the whole Riemann sphere, but also many maps that are not conjugate to any rational map. This makes them a natural setting for exploring connections between dynamical systems, fractal geometry, and algebra. Every expanding Thurston map gives rise to two seemingly unrelated objects. The first is its visual sphere: the underlying 2-sphere equipped with a fractal metric in which the map uniformly expands distances at small scales. The second is its iterated monodromy group: a self-similar group acting on the infinite rooted tree of iterated preimages of a basepoint. In this talk, I will introduce both objects through explicit examples. I will then discuss recent developments and open questions concerning the interplay between the dynamics of expanding Thurston maps, the geometry of their visual spheres, and the amenability properties of their iterated monodromy groups.
We introduce a computable analysis theory for Markov partitions which induces computationally effective symbolic-dynamic machinery for certain topological dynamical systems. In this theory, Markov partitions are represented by families of computable subsets of a computable metric space. Once Markov partitions are computable objects, classical symbolic reductions to shift spaces can be computed algorithmically. With these reductions, we are able to compute many invariants of topological dynamical systems that were previously inaccessible. We illustrate our techniques on fundamental examples of dynamical systems admitting Markov partitions, such as distance expanding maps on a circle and limit sets of iterated function systems. In these examples, we not only show how to construct Markov partitions at any radius but also demonstrate how to apply these partitions to compute invariants such as the topological pressure of continuous potentials, equilibrium states, zero-temperature measures, and the Hausdorff dimension of limit sets of iterated function systems.
Bistable genetic regulatory networks (GRNs) are ubiquitous in living organisms. Understanding how evolution has tuned the stability of biological switches is an important problem. This understanding can guide the design of forward-engineered synthetic bistable GRNs. Since producing functional regulator proteins takes time, delay is an inherent feature of GRNs. This delay is distributed (random) because protein production is inherently noisy. We examine how distributed delay impacts the dynamics of bistable GRNs.
We consider the number of hits along the trajectory in Kac's time in a random skew product dynamical system with certain geometric properties. We prove that the quenched measure of the number of the hits will have a Poisson limit for almost every $(\omega,x) \in \Omega \times M$.