BEGIN:VCALENDAR VERSION:2.0 PRODID:-//The University of Alabama//Applied Mathematics Seminar build.py// EN CALSCALE:GREGORIAN METHOD:PUBLISH X-WR-CALNAME:UA Applied Mathematics Seminar X-WR-TIMEZONE:America/Chicago BEGIN:VEVENT UID:20260911-applied-math-seminar@jiazhaomath.github.io DTSTAMP:20261007T000000Z DTSTART:20260911T160000Z DTEND:20260911T165000Z SUMMARY:Applied Mathematics Seminar: Matthias Dogbatsey — A Diffuse-Inter face Variational Implicit Solvation Model with the Size-Modified Poisson-B oltzmann Equation: Formulation\, Numerical Computation\, and Stern-Layer F ormation LOCATION:GP 346\, The University of Alabama\, Tuscaloosa\, Alabama DESCRIPTION:Matthias Dogbatsey\, PhD student\, Department of Mathematics\, The University of Alabama\nA Diffuse-Interface Variational Implicit Solvat ion Model with the Size-Modified Poisson-Boltzmann Equation: Formulation\, Numerical Computation\, and Stern-Layer Formation\nDiffuse-interface vari ational implicit solvation models (VISMs) enable efficient computation of ensemble average solvation energies without costly molecular dynamics samp ling. However\, the underlying coupled nonlinear PDEs remain computational ly demanding\, and the classical Poisson–Boltzmann model neglects finite ion size\, predicting unphysically high concentrations near molecular sur faces. We present the first numerical realization of the diffuse-interface VISM\, developed to date only analytically\, and the first to incorporate steric effects via the fully nonlinear size-modified Poisson–Boltzmann (SMPB) equation\, coupled to a p-Laplace equation that determines the solu te-solvent interface variationally rather than prescribing it. Our solver combines an alternating direction implicit scheme with an inexact Newton –Krylov method\, stabilized by a coupling relaxation strategy. We valida te against manufactured solutions and biomolecular benchmarks\, and show t hat both steric ion effects and short-range Lennard-Jones repulsion are re quired for a Stern layer to emerge.\nhttps://jiazhaomath.github.io/seminar /2026-2027.html#2026-09-11 URL:https://jiazhaomath.github.io/seminar/2026-2027.html#2026-09-11 END:VEVENT BEGIN:VEVENT UID:20260925-applied-math-seminar@jiazhaomath.github.io DTSTAMP:20261007T000000Z DTSTART:20260925T160000Z DTEND:20260925T165000Z SUMMARY:Applied Mathematics Seminar: Zheng Sun — OEDG: Oscillation-Elimin ating Discontinuous Galerkin Method for Hyperbolic Conservation Laws LOCATION:GP 346\, The University of Alabama\, Tuscaloosa\, Alabama DESCRIPTION:Zheng Sun\, Associate Professor\, Department of Mathematics\, T he University of Alabama\nOEDG: Oscillation-Eliminating Discontinuous Gale rkin Method for Hyperbolic Conservation Laws\nControlling spurious oscilla tions is crucial for designing reliable numerical schemes for hyperbolic c onservation laws. In this talk\, we propose a novel\, robust\, and efficie nt oscillation-eliminating discontinuous Galerkin (OEDG) method on general meshes\, motivated by the damping technique in [Lu\, Liu\, and Shu\, SIAM J. Numer. Anal.\, 59:1299–1324\, 2021]. The OEDG method incorporates an OE procedure after each Runge–Kutta stage\, and it is devised by altern ately evolving the conventional semidiscrete DG scheme and a damping equat ion. A novel damping operator is designed to possess both scale-invariant and evolution-invariant properties. We rigorously prove optimal error esti mates of the fully discrete OEDG method for smooth solutions of linear sca lar conservation laws. The OEDG method exhibits many notable advantages. I t effectively eliminates spurious oscillations for problems spanning vario us scales and wave speeds without problem-specific parameters. Furthermore \, it retains the key properties of the conventional DG method\, such as c onservation\, optimal convergence rates\, and superconvergence. Moreover\, the OEDG method maintains stability under the normal CFL condition\, even in the presence of strong shocks associated with highly stiff damping ter ms. The OE procedure is non-intrusive\, facilitating seamless integration into existing DG codes as an independent module. Extensive numerical resul ts confirm the analysis and validate the effectiveness of the OEDG method. \nhttps://jiazhaomath.github.io/seminar/2026-2027.html#2026-09-25 URL:https://jiazhaomath.github.io/seminar/2026-2027.html#2026-09-25 END:VEVENT BEGIN:VEVENT UID:20261002-applied-math-seminar@jiazhaomath.github.io DTSTAMP:20261007T000000Z DTSTART:20261002T160000Z DTEND:20261002T165000Z SUMMARY:Applied Mathematics Seminar: Keren Li — Representative Learning: From Point Representatives to Distributed AI LOCATION:GP 346\, The University of Alabama\, Tuscaloosa\, Alabama DESCRIPTION:Keren Li\, Assistant Professor\, Department of Mathematics\, Un iversity of Alabama at Birmingham\nRepresentative Learning: From Point Rep resentatives to Distributed AI\nDistributed machine learning traditionally relies on exchanging gradients\, model parameters\, or predictions across computing nodes. Such communication can become inefficient and unstable w hen data are heterogeneous or communication is asynchronous. In this talk\ , I present Representative Learning (RepL)\, a framework that replaces mod el-based communication with the exchange of representative data objects th at summarize local statistical information while preserving privacy and in terpretability.\n\nI will describe the evolution of RepL from simple point representatives to score-preserving and distribution-valued representatio ns\, highlighting recent developments in stability\, identifiability\, and heterogeneity-aware learning. I will also discuss extensions to collabora tive learning\, clouded representatives based on lifted empirical risk\, a nd distributed AI systems that communicate information at the data level r ather than the model level. The talk concludes with current challenges and future directions toward a unified framework for distributed AI under het erogeneity and communication constraints.\nhttps://jiazhaomath.github.io/s eminar/2026-2027.html#2026-10-02 URL:https://jiazhaomath.github.io/seminar/2026-2027.html#2026-10-02 END:VEVENT BEGIN:VEVENT UID:20261009-applied-math-seminar@jiazhaomath.github.io DTSTAMP:20261007T000000Z DTSTART:20261009T160000Z DTEND:20261009T165000Z SUMMARY:Applied Mathematics Seminar: Baoli Hao — What Can Ionic Transport Data Reveal About Steric Interactions? LOCATION:GP 346\, The University of Alabama\, Tuscaloosa\, Alabama DESCRIPTION:Baoli Hao\, Ph.D. candidate in Applied Mathematics\, Illinois I nstitute of Technology\nWhat Can Ionic Transport Data Reveal About Steric Interactions?\nTo what extent can microscopic interactions be inferred fro m observable dynamics? We investigate this question in ionic transport\, w here steric interactions may operate at scales beyond experimental resolut ion. Within a unified energy-dissipation framework\, we develop a hierarch y of local\, higher-order\, and nonlocal constitutive models to investigat e what can be learned from dynamical observations. By connecting interacti on scales and observational information with identifiability\, model stabi lity\, and predictive accuracy\, we determine the level of constitutive co mplexity supported by the data. Our study highlights the distinction betwe en recovering microscopic interactions and identifying their observable dy namical effects\, providing a general perspective on data-driven coarse-gr aining and the limits of learning multiscale physics from finite-resolutio n observations.\nhttps://jiazhaomath.github.io/seminar/2026-2027.html#2026 -10-09 URL:https://jiazhaomath.github.io/seminar/2026-2027.html#2026-10-09 END:VEVENT BEGIN:VEVENT UID:20261016-applied-math-seminar@jiazhaomath.github.io DTSTAMP:20261007T000000Z DTSTART:20261016T150000Z DTEND:20261016T155000Z SUMMARY:Applied Mathematics Seminar: Aina G. Irbe — Towards Inclusive Int elligence: Defining the Research Challenges at the Intersection of Artific ial Intelligence\, Universal Design for Learning\, and Workplace Training LOCATION:GP 208\, The University of Alabama\, Tuscaloosa\, Alabama DESCRIPTION:Aina G. Irbe\, Chief Operating Officer\, Accessible Minds\nTowa rds Inclusive Intelligence: Defining the Research Challenges at the Inters ection of Artificial Intelligence\, Universal Design for Learning\, and Wo rkplace Training\nEarlier time and different room: 10:00–10:50 AM in GP 208.\nOrganizations are increasingly using artificial intelligence (AI) to create workplace and academic training. AI can generate content quickly\, adapt materials to different learners\, and deliver information in text\, audio\, and video. These features seem to align with Universal Design for Learning (UDL)\, a framework that supports diverse learners by providing multiple means of engagement\, representation\, and action and expression. However\, being efficient and adaptive is not the same as being inclusive . AI-generated training can still overlook barriers related to disability\ , accessibility\, language\, cognition\, and knowledge transfer.\n\nThis r aises three challenges in using AI to develop workplace and academic train ing: AI is often optimized for goals that don't reflect inclusive learning \, it can introduce new accessibility barriers or amplify existing ones\, and there is no practical\, validated way to evaluate training developed w ith AI. A proposed six-dimension framework based on UDL and the Web Conten t Accessibility Guidelines (WCAG) offers a way to assess whether AI-develo ped training is accessible\, inclusive\, and educationally sound. Three re search questions and recommendations for future studies outline the next s teps needed to validate it.\nhttps://jiazhaomath.github.io/seminar/2026-20 27.html#2026-10-16 URL:https://jiazhaomath.github.io/seminar/2026-2027.html#2026-10-16 END:VEVENT BEGIN:VEVENT UID:20261106-applied-math-seminar@jiazhaomath.github.io DTSTAMP:20261007T000000Z DTSTART:20261106T170000Z DTEND:20261106T175000Z SUMMARY:Applied Mathematics Seminar: Paul Laiu — Title TBA DESCRIPTION:Paul Laiu\, Staff Mathematician\, Oak Ridge National Laboratory \nTitle TBA\nhttps://jiazhaomath.github.io/seminar/2026-2027.html#2026-11- 06 URL:https://jiazhaomath.github.io/seminar/2026-2027.html#2026-11-06 END:VEVENT BEGIN:VEVENT UID:20261113-applied-math-seminar@jiazhaomath.github.io DTSTAMP:20261007T000000Z DTSTART:20261113T170000Z DTEND:20261113T175000Z SUMMARY:Applied Mathematics Seminar: Vadim Zharnitsky — Optimal damping i n the mass-spring systems DESCRIPTION:Vadim Zharnitsky\, Professor\, Department of Mathematics\, Univ ersity of Illinois Urbana-Champaign\nOptimal damping in the mass-spring sy stems\nIt is a well-known textbook example in classical mechanics that a s ingle-degree-of-freedom mass-spring harmonic oscillator achieves its faste st exponential decay when the damping is critical. We study the analogous problem of designing optimal damping for a system of several coupled linea r oscillators. Using the Weyl–Horn theorem\, we construct an explicit da mping matrix that theoretically realizes the best possible exponential dec ay rate of the system\, which turns out to be the geometric mean of the na tural frequencies. We show that this damping strategy outperforms the clas sical Rayleigh (proportional) damping approach. We also characterize the p arameter regimes in which the optimal damping matrix necessarily ceases to be positive definite\, and discuss the implications of this phenomenon.\n \nThis is a joint work with M. Arnold (UT-Dallas\, USA) and O. Gendelman ( Technion\, Israel).\nhttps://jiazhaomath.github.io/seminar/2026-2027.html# 2026-11-13 URL:https://jiazhaomath.github.io/seminar/2026-2027.html#2026-11-13 END:VEVENT BEGIN:VEVENT UID:20261120-applied-math-seminar@jiazhaomath.github.io DTSTAMP:20261007T000000Z DTSTART:20261120T170000Z DTEND:20261120T175000Z SUMMARY:Applied Mathematics Seminar: Kirill Golubnichiy — A Semi-Discrete Convexification Method Combined with Deep Learning for Electrical Impedan ce Tomography DESCRIPTION:Kirill Golubnichiy\, Postdoc\, Department of Mathematics & Stat istics\, Texas Tech University\nA Semi-Discrete Convexification Method Com bined with Deep Learning for Electrical Impedance Tomography\nIn this talk \, we develop a new semi-discrete version of the Carleman estimate-based c onvexification method\, a globally convergent numerical approach for nonli near inverse problems. The semi-discrete convexification method is used to generate an accurate starting point for the subsequent deep learning trai ning procedure. An important feature of the continuous convexification met hod is that its convergence to the true solution does not require the avai lability of a sufficiently accurate initial guess.\n\nWe introduce a new c oncept of (h)-strong convexity\, where (h) denotes the grid step size in t he semi-discrete formulation. This property enables us to derive an a prio ri accuracy estimate for the starting point supplied to the deep learning stage. The proposed approach is applied to the highly nonlinear inverse pr oblem of Electrical Impedance Tomography. Numerical experiments involving complicated conductivity structures demonstrate the accuracy\, robustness\ , and computational feasibility of the combined semi-discrete convexificat ion and deep learning procedure.\nhttps://jiazhaomath.github.io/seminar/20 26-2027.html#2026-11-20 URL:https://jiazhaomath.github.io/seminar/2026-2027.html#2026-11-20 END:VEVENT BEGIN:VEVENT UID:20261204-applied-math-seminar@jiazhaomath.github.io DTSTAMP:20261007T000000Z DTSTART:20261204T170000Z DTEND:20261204T175000Z SUMMARY:Applied Mathematics Seminar: Chunyan Li — Title TBA DESCRIPTION:Chunyan Li\, Postdoc\, Department of Mathematics\, University o f South Carolina\nTitle TBA\nhttps://jiazhaomath.github.io/seminar/2026-20 27.html#2026-12-04 URL:https://jiazhaomath.github.io/seminar/2026-2027.html#2026-12-04 END:VEVENT END:VCALENDAR