This paper provides a proposed means to estimate parameters of noise corrupted oscillator systems. An application for a submarine combat control systems (CCS) rack is described as exemplary of the method.

Francis J. OBrien Jr Nathan Johnnie Susan Maloney Aimee Ross

We show that linear complexity is the threshold for the emergence of Kakutani inequivalence for measurable systems supported on a minimal subshift. In particular, we show that there are minimal subshifts of arbitrarily low super-linear complexity that admit both loosely Bernoulli and non-loosely Bernoulli ergodic measures and that no minimal subshift with linear complexity can admit inequivalent measures.

Van Cyr Aimee Johnson Bryna Kra Ayse Sahin

We study minimal $\mathbb{Z}^d$-Cantor systems and the relationship between their speedups, their collections of invariant Borel measures, their associated unital dimension groups, and their orbit equivalence classes. In the particular case of minimal $\mathbb{Z}^d$-odometers, we show that their bounded speedups must again be odometers but, contrary to the 1-dimensional case, they need not be conjugate, or even isomorphic, to the original.

Aimee S. A. Johnson David M. McClendon

Compton scattering within the accretion column of magnetic cataclysmic variables (mCVs) can induce a net polarization in the X-ray emission. We investigate this process using Monte Carlo simulations and find that significant polarization can arise as a result of the stratified flow structure in the shock-ionized column. We find that the degree of linear polarization can reach levels up to ~8% for systems with high accretion rates and low white-dwarf masses, when viewed at large inclination angles with respect to the accretion column axis. These levels are substantially higher than previously predicted estimates using an accretion column model with uniform density and temperature. We also find that for systems with a relatively low-mass white dwarf accreting at a high accretion rate, the polarization properties may be insensitive to the magnetic field, since most of the scattering occurs at the base of the accretion column where the density structure is determined mainly by bremsstrahlung cooling instead of cyclotron cooling.

Aimee McNamara Zdenka Kuncic Kinwah Wu

We introduce a multi-fidelity estimator of covariance matrices that employs the log-Euclidean geometry of the symmetric positive-definite manifold. The estimator fuses samples from a hierarchy of data sources of differing fidelities and costs for variance reduction while guaranteeing definiteness, in contrast with previous approaches. The new estimator makes covariance estimation tractable in applications where simulation or data collection is expensive; to that end, we develop an optimal sample allocation scheme that minimizes the mean-squared error of the estimator given a fixed budget. Guaranteed definiteness is crucial to metric learning, data assimilation, and other downstream tasks. Evaluations of our approach using data from physical applications (heat conduction, fluid dynamics) demonstrate more accurate metric learning and speedups of more than one order of magnitude compared to benchmarks.

Aimee Maurais Terrence Alsup Benjamin Peherstorfer Youssef Marzouk

We introduce a multifidelity estimator of covariance matrices formulated as the solution to a regression problem on the manifold of symmetric positive definite matrices. The estimator is positive definite by construction, and the Mahalanobis distance minimized to obtain it possesses properties enabling practical computation. We show that our manifold regression multifidelity (MRMF) covariance estimator is a maximum likelihood estimator under a certain error model on manifold tangent space. More broadly, we show that our Riemannian regression framework encompasses existing multifidelity covariance estimators constructed from control variates. We demonstrate via numerical examples that the MRMF estimator can provide significant decreases, up to one order of magnitude, in squared estimation error relative to both single-fidelity and other multifidelity covariance estimators. Furthermore, preservation of positive definiteness ensures that our estimator is compatible with downstream tasks, such as data assimilation and metric learning, in which this property is essential.

Aimee Maurais Terrence Alsup Benjamin Peherstorfer Youssef Marzouk

We introduce a new mean-field ODE and corresponding interacting particle systems (IPS) for sampling from an unnormalized target density. The IPS are gradient-free, available in closed form, and only require the ability to sample from a reference density and compute the (unnormalized) target-to-reference density ratio. The mean-field ODE is obtained by solving a Poisson equation for a velocity field that transports samples along the geometric mixture of the two densities, which is the path of a particular Fisher-Rao gradient flow. We employ a RKHS ansatz for the velocity field, which makes the Poisson equation tractable and enables discretization of the resulting mean-field ODE over finite samples. The mean-field ODE can be additionally be derived from a discrete-time perspective as the limit of successive linearizations of the Monge-Amp\`ere equations within a framework known as sample-driven optimal transport. We introduce a stochastic variant of our approach and demonstrate empirically that our IPS can produce high-quality samples from varied target distributions, outperforming comparable gradient-free particle systems and competitive with gradient-based alternatives.

Aimee Maurais Youssef Marzouk

We have characterized the two-dimensional spatial dependence of the hydrodynamic interactions between two adhesively rolling leukocytes in a live venule in the mouse cremaster muscle. Two rolling leukocytes were observed to slow each other down when rolling together in close proximity, due to mutual sheltering from the external blood flow in the vessel lumen. These results are in agreement with a previous study of leukocyte rolling interactions using carbohydrate-coated beads in a parallel-plate flow chamber and a detailed computer model of adhesion in a multicellular environment.

Michael R. King Aimee D. Ruscio Michael B. Kim Ingrid H. Sarelius

We investigated the effects of student-generated problems on exams. The process was gradual with some training throughout the semester. Initial results were highly positive with the students involved performing significantly better, and showing statistically significant improvement (t = 5.04) compared to the rest of the class, on average. Overall, performance improved when students generated problems. Motivation was a limiting factor. There is significant potential for improving student learning of physics and other problem-based topics.

Ameya S. Kolarkar Aimee A. Callender

In this paper, we give explicit conditions characterizing the F{\o}lner rank one $\mathbb{Z}^d$-actions that factor onto a finite odometer; those that factor onto an arbitrary, but specified $\mathbb{Z}^d$-odometer, and those that factor onto an unspecified $\mathbb{Z}^d$-odometer. We also give explicit conditions describing the F{\o}lner rank one $\mathbb{Z}^d$-actions that are conjugate to a specific $\mathbb{Z}^d$-odometer, and those that are conjugate to some $\mathbb{Z}^d$-odometer. These conditions are based on cutting and stacking procedures used to generate the action, and generalize results given in \cite{FGHSW} for rank one $\mathbb{Z}$-actions.

Aimee S. A. Johnson David M. McClendon