This Week in Mathematics


>> Next Week [2025-04-06 - 2025-04-12]	>> Beyond Next Week [2025-04-12+]

<< View Previously Scheduled Events

Current Week [Mar 30, 2025 - Apr 05, 2025]

Next MonthApr

Mar - Apr
S	M	T	W	R	F	S
						1
2	3	4	5	6	7	8
9	10	11	12	13	14	15
16	17	18	19	20	21	22
23	24	25	26	27	28	29
30	31	1	2	3	4	5

Today:

Analysis and PDE Seminar
Self-similar instability and forced nonuniqueness
- Michele Dolce, EPFL, Switzerland
Time: 3.05pm Room: LOV 232
More Information
Abstract/Desc: Building on an approach introduced by Golovkin in the '60s, we show that nonuniqueness in some forced PDEs is a direct consequence of the existence of a self-similar linearly unstable eigenvalue: the key point is a clever choice of the forcing term removing complicated nonlinear interactions. We use this method to present a short and self-contained proof of nonuniqueness in 2D perfect fluids, first obtained in Vishik's groundbreaking result. In particular, we give a direct construction of a forced self-similar unstable vortex, where we treat perturbatively the self-similar operator in a new and more quantitative way. This is a joint work with Giulia Mescolini.

Entries for this week: 6

Monday March 31, 2025

Analysis and PDE Seminar
Self-similar instability and forced nonuniqueness
- Michele Dolce, EPFL, Switzerland
Time: 3.05pm Room: LOV 232
More Information
Abstract/Desc: Building on an approach introduced by Golovkin in the '60s, we show that nonuniqueness in some forced PDEs is a direct consequence of the existence of a self-similar linearly unstable eigenvalue: the key point is a clever choice of the forcing term removing complicated nonlinear interactions. We use this method to present a short and self-contained proof of nonuniqueness in 2D perfect fluids, first obtained in Vishik's groundbreaking result. In particular, we give a direct construction of a forced self-similar unstable vortex, where we treat perturbatively the self-similar operator in a new and more quantitative way. This is a joint work with Giulia Mescolini.

Tuesday April 01, 2025

ACM seminar
Conditional neural field latent diffusion modeling for spatiotemporal turbulence modeling
- Meet Parikh, University of Notre Dame
Time: 3:05 pm Room: 0231
Abstract/Desc: Simulating complex, unsteady turbulent flows with high fidelity is critical across science and engineering, yet traditional methods like Direct Numerical Simulation (DNS) and Large Eddy Simulation (LES) face significant computational constraints, while many deep learning surrogates struggle to capture the inherent stochasticity and complexity, particularly within irregular geometries or under varying conditions. To overcome these hurdles, we introduce the Conditional Neural Field Latent Diffusion (CoNFiLD) model, a generative framework leveraging the synergy between conditional neural fields and latent diffusion processes. This approach enables efficient, memory-conscious, and robust stochastic generation of spatiotemporal turbulent flows, flexibly adapting to diverse physical parameters and boundary conditions without retraining, facilitating applications like sparse data reconstruction and super-resolution. A particularly challenging application demanding such capabilities is the synthesis of realistic, stochastic inflow conditions for eddy-resolving simulations, where conventional recycling methods are prohibitively expensive and existing synthetic or deterministic generators often fail to replicate key turbulence structures or maintain long-term stability. The CoNFiLD framework directly addresses this need through CoNFiLD-inlet, a specialized inflow generator. Demonstrating the power of conditional generation, CoNFiLD-inlet effectively generalizes across a wide range of Reynolds numbers using a single trained model. Comprehensive validation through a-priori and a-posteriori tests within DNS and Wall-Modeled Large Eddy Simulation (WMLES) substantiates the high fidelity, robustness, and scalability of this approach. This successful application underscores CoNFiLD’s potential as a versatile, computationally efficient tool poised to advance real-time turbulence simulation and enable digital twin technologies in fluid dynamics.

Geometry and Topology Seminar [url]
Applications of nonstandard analysis in geometry
- Sam Ballas, FSU
Time: 3:05 Room: 232
More Information
Abstract/Desc: Non standard analysis is a version of analysis where the reals are replaced by the larger non-archimedian field of hyperreals, which contain infinitesimally small elements. In this talk I will introduce the hyperreals and discuss how they give a useful language for describing various degeneration phenomena in geometry.

Wednesday April 02, 2025

Biomathematics Journal Club
A Unified Mathematical Model of Thyroid Hormone Regulation and Implication for Personalized Treatment of Thyroid Disorders
- Sarah Romero, FSU
Time: 5:00 Room: Dirac Library

Thursday April 03, 2025

Algebra seminar
Toric Geometry and Matroid Chern-Schwartz-MacPherson Classes
- Jeffery Liu, FSU
Time: 3:05PM Room: 232
Abstract/Desc: Toric varieties are a type of algebraic variety containing the algebraic torus as a dense open subset. There is a connection between toric varieties and rational polyhedral fans. The Chow cohomology ring of a toric variety can be deduced from the geometry of a corresponding fan. Toric vector bundles are vector bundles on a toric variety with a torus action and have Chern classes valued in its Chow ring. Matroids are a combinatorial object that encodes “dependence properties” among elements of a set; for example, linear dependence in a vector frame. Associated to each matroid is a polyhedral fan called the Bergman fan. The matroid has a characteristic class called the Chern-Schwartz-MacPherson (CSM) class which is the Chern class of a “tautological” sub-bundle associated to the matroid. This class can be used, for example, to distinguish when two matroids are non-isomorphic. We give a basic overview of the theory of toric varieties, polyhedral fans, and toric vector bundles, and use it to compute the CSM class of a matroid.

Financial Mathematics Seminar
- Ruth Lopez Fajardo, FSU
Time: 3:05 Room: LOV 0231

Problems? Email webmaster@math.fsu.edu.

Department of Mathematics

This Week in Mathematics