Asymptotics of blowup solutions for the aggregation equation
Yanghong Huang Andrea Bertozzi
We consider the asymptotic behavior of radially symmetric solutions of the aggregation equation $ u_t = \nabla\cdot(u\nabla K*u) $ in $\mathbb{R}^n$, for homogeneous potentials $K(x) = |x|^\gamma$, $\gamma>0$. For $\gamma>2$, the aggregation happens in infinite time and exhibits a concentration of mass along a collapsing $\delta$-ring. We develop an asymptotic theory for the approach to this singular solution. For $\gamma < 2$, the solution blows up in finite time and we present careful numerics of second type similarity solutions for all $\gamma$ in this range, including additional asymptotic behaviors in the limits $\gamma \to 0^+$ and $\gamma\to 2^-$.
keywords: asymptotic behavior self-similar solutions. blowup Aggregation equation
Thomas P. Witelski David M. Ambrose Andrea Bertozzi Anita T. Layton Zhilin Li Michael L. Minion
Studies of problems in fluid dynamics have spurred research in many areas of mathematics, from rigorous analysis of nonlinear partial differential equations, to numerical analysis, to modeling and applied analysis of related physical systems. This special issue of Discrete and Continuous Dynamical Systems Series B is dedicated to our friend and colleague Tom Beale in recognition of his important contributions to these areas.

For more information please click the "Full Text" above.
Two-species particle aggregation and stability of co-dimension one solutions
Alan Mackey Theodore Kolokolnikov Andrea L. Bertozzi
Systems of pairwise-interacting particles model a cornucopia of physical systems, from insect swarms and bacterial colonies to nanoparticle self-assembly. We study a continuum model with densities supported on co-dimension one curves for two-species particle interaction in $\mathbb{R}^2$, and apply linear stability analysis of concentric ring steady states to characterize the steady state patterns and instabilities which form. Conditions for linear well-posedness are determined and these results are compared to simulations of the discrete particle dynamics, showing predictive power of the linear theory. Some intriguing steady state patterns are shown through numerical examples.
keywords: measure solutions Aggregation equation continuum limit linear well-posedness linear stability asymptotics. pattern formation
Positive feedback control of Rayleigh-Bénard convection
B. A. Wagner Andrea L. Bertozzi L. E. Howle
We consider the problem of active feedback control of Rayleigh-Bénard convection via shadowgraphic measurement. Our theoretical studies show, that when the feedback control is positive, i.e. is tuned to advance the onset of convection, there is a critical threshold beyond which the system becomes linearly ill-posed so that short-scale disturbances are greatly amplified. Experimental observation suggests that finite size effects become important and we develop a theory to explain these contributions. As an efficient modelling tool for studying the dynamics of such a controlled pattern forming system, we use a Galerkin approximation to derive a dimension reduced model.
keywords: control Galerkin projection illposedness. Rayleigh-Bénard convection
Existence and uniqueness of solutions to an aggregation equation with degenerate diffusion
Andrea L. Bertozzi Dejan Slepcev
We present an energy-methods-based proof of the existence and uniqueness of solutions of a nonlocal aggregation equation with degenerate diffusion. The equation we study is relevant to models of biological aggregation.
keywords: Aggregation nonlocal interaction.
Cops on the dots in a mathematical model of urban crime and police response
Joseph R. Zipkin Martin B. Short Andrea L. Bertozzi
Hotspots of crime localized in space and time are well documented. Previous mathematical models of urban crime have exhibited these hotspots but considered a static or otherwise suboptimal police response to them. We introduce a program of police response to hotspots of crime in which the police adapt dynamically to changing crime patterns. In particular, they choose their deployment to solve an optimal control problem at every time. This gives rise to a free boundary problem for the police deployment's spatial support. We present an efficient algorithm for solving this problem numerically and show that police presence can prompt surprising interactions among adjacent hotspots.
keywords: free boundary problem Reaction-diffusion equations crime modeling. optimal control
Preface to special issue on mathematics of social systems
Andrea L. Bertozzi
This special issue is an outgrowth of a minisyposium titled ``Mathematics of Social Systems" held at the 9th AIMS Conference on Dynamical Systems, Differential Equations and Applications, held in Orlando, FL in July 2012. Presenters from that session were invited to submit papers that were reviewed using the usual procedures of the DCDS journals, along with additional authors from the field. Mathematics has already had a significant impact on basic research involving fundamental problems in physical sciences, biological sciences, computer science and engineering. Examples include understanding of the equations of incompressible fluid dynamics, shock wave theory and compressible gas dynamics, ocean modeling, algorithms for image processing and compressive sensing, and biological problems such as models for invasive species, spread of disease, and more recently systems biology for modeling of complex organisms and complex patterns of disease. This impact has yet to come to fruition in a comprehensive way for complex social behavior. While computational models such as agent-based systems and well-known statistical methods are widely used in the social sciences, applied mathematics has not to date had a core impact in the social sciences at the level that it achieves in the physical and life sciences. However in recent years we have seen a growth of work in this direction and ensuing new mathematics problems that must be tackled to understand such problems. Technical approaches include ideas from statistical physics, nonlinear partial differential equations of all types, statistics and inverse problems, and stochastic processes and social network models. The collection of papers presented in this issue provides a backdrop of the current state of the art results in this developing new research area in applied mathematics. The body of work encompasses many of the challenges in understanding these discrete complex systems and their related continuum approximations.

For more information please click the “Full Text” above.
keywords: pattern formation agent based model Social Sciences network analysis stochastics.
A biharmonic-modified forward time stepping method for fourth order nonlinear diffusion equations
Andrea L. Bertozzi Ning Ju Hsiang-Wei Lu
We consider a class of splitting schemes for fourth order nonlinear diffusion equations. Standard backward-time differencing requires the solution of a higher order elliptic problem, which can be both computationally expensive and work-intensive to code, in higher space dimensions. Recent papers in the literature provide computational evidence that a biharmonic-modified, forward time-stepping method, can provide good results for these problems. We provide a theoretical explanation of the results. For a basic nonlinear 'thin film' type equation we prove $H^1$ stability of the method given very simple boundedness constraints of the numerical solution. For a more general class of long-wave unstable problems, we prove stability and convergence, using only constraints on the smooth solution. Computational examples include both the model of 'thin film' type problems and a quantitative model for electrowetting in a Hele-Shaw cell (Lu et al J. Fluid Mech. 2007). The methods considered here are related to 'convexity splitting' methods for gradient flows with nonconvex energies.
keywords: Fourth order equations convex splitting.
$H^1$ Solutions of a class of fourth order nonlinear equations for image processing
John B. Greer Andrea L. Bertozzi
Recently fourth order equations of the form $u_t = -\nabla\cdot((\mathcal G(J_\sigma u)) \nabla \Delta u)$ have been proposed for noise reduction and simplification of two dimensional images. The operator $\mathcal G$ is a nonlinear functional involving the gradient or Hessian of its argument, with decay in the far field. The operator $J_\sigma$ is a standard mollifier. Using ODE methods on Sobolev spaces, we prove existence and uniqueness of solutions of this problem for $H^1$ initial data.
keywords: image processing ODEs on a Banach space mollifiers anisotropic diffusion Fourth order partial differential equations higher order nonlinear PDEs.
Image segmentation with dynamic artifacts detection and bias correction
Dominique Zosso Jing An James Stevick Nicholas Takaki Morgan Weiss Liane S. Slaughter Huan H. Cao Paul S. Weiss Andrea L. Bertozzi
Region-based image segmentation is well-addressed by the Chan-Vese (CV) model. However, this approach fails when images are affected by artifacts (outliers) and illumination bias that outweigh the actual image contrast. Here, we introduce a model for segmenting such images. In a single energy functional, we introduce 1) a dynamic artifact class preventing intensity outliers from skewing the segmentation, and 2), in Retinex-fashion, we decompose the image into a piecewise-constant structural part and a smooth bias part. The CV-segmentation terms then only act on the structure, and only in regions not identified as artifacts. The segmentation is parameterized using a phase-field, and efficiently minimized using threshold dynamics.We demonstrate the proposed model on a series of sample images from diverse modalities exhibiting artifacts and/or bias. Our algorithm typically converges within 10-50 iterations and takes fractions of a second on standard equipment to produce meaningful results. We expect our method to be useful for damaged images, and anticipate use in applications where artifacts and bias are actual features of interest, such as lesion detection and bias field correction in medical imaging, e.g., in magnetic resonance imaging (MRI).
keywords: Image segmentation artifacts bias Retinex dynamic labeling ChanVese levelset method threshold dynamics MBO

Year of publication

Related Authors

Related Keywords

[Back to Top]