Universality in the structure of dark matter haloes over twenty orders of magnitude in halo mass. (arXiv:1911.09720v1 [astro-ph.CO])

Dark matter haloes are the basic units of all cosmic structure. They grew by gravitational amplification of weak initial density fluctuations that are still visible on large scales in the cosmic microwave background radiation. Galaxies formed within relatively massive haloes as gas cooled and condensed at their centres, but many hypotheses for the nature of…

On the closure requirement for VOF simulations with RANS modeling. (arXiv:1911.09727v1 [physics.comp-ph])

The volume of fluid (VOF) method is increasingly used in computational fluid dynamics (CFD) simulations of turbulent two-phase flows. The Reynolds-Averaged Navier-Stokes (RANS) approach is an economic and practical way for turbulent VOF simulations. Even though RANS-VOF simulations are widely conducted, the underlying physics is barely discussed. This study reveals the very basic closure requirement…

Quantum Lissajous Scars. (arXiv:1911.09729v1 [quant-ph])

A quantum scar – an enhancement of a quantum probability density in the vicinity of a classical periodic orbit – is a fundamental phenomenon connecting quantum and classical mechanics. Here we demonstrate that some of the eigenstates of the perturbed two-dimensional anisotropic (elliptic) harmonic oscillator are strongly scarred by the Lissajous orbits of the unperturbed…

Filter Response Normalization Layer: Eliminating Batch Dependence in the Training of Deep Neural Networks. (arXiv:1911.09737v1 [cs.LG])

Batch Normalization (BN) is a highly successful and widely used batch dependent training method. Its use of mini-batch statistics to normalize the activations introduces dependence between samples, which can hurt the training if the mini-batch size is too small, or if the samples are correlated. Several alternatives, such as Batch Renormalization and Group Normalization (GN),…

Controlling False Discovery Rate Using Gaussian Mirrors. (arXiv:1911.09761v1 [stat.ME])

Simultaneously finding multiple influential variables and controlling the false discovery rate (FDR) for linear regression models is a fundamental problem with a long history. We here propose the Gaussian Mirror (GM) method, which creates for each predictor variable a pair of mirror variables by adding and subtracting a randomly generated Gaussian random variable, and proceeds…

Mixture survival models methodology: an application to cancer immunotherapy assessment in clinical trials. (arXiv:1911.09765v1 [stat.AP])

Progress in immunotherapy revolutionized the treatment landscape for advanced lung cancer, raising survival expectations beyond those that were historically anticipated with this disease. In the present study, we describe the methods for the adjustment of mixture parametric models of two populations for survival analysis in the presence of long survivors. A methodology is proposed in…

Geo-clustered chronic affinity: pathways from socio-economic disadvantages to health disparities. (arXiv:1911.09769v1 [cs.CY])

Our objective was to develop and test a new concept (affinity) analogous to multimorbidity of chronic conditions for individuals at census tract level in Memphis, TN. The use of affinity will improve the surveillance of multiple chronic conditions and facilitate the design of effective interventions. We used publicly available chronic condition data (Center for Disease…

Poisson-Minibatching for Gibbs Sampling with Convergence Rate Guarantees. (arXiv:1911.09771v1 [cs.LG])

Gibbs sampling is a Markov chain Monte Carlo method that is often used for learning and inference on graphical models. Minibatching, in which a small random subset of the graph is used at each iteration, can help make Gibbs sampling scale to large graphical models by reducing its computational cost. In this paper, we propose…

Effects of Differential Privacy and Data Skewness on Membership Inference Vulnerability. (arXiv:1911.09777v1 [cs.CR])

Membership inference attacks seek to infer the membership of individual training instances of a privately trained model. This paper presents a membership privacy analysis and evaluation system, called MPLens, with three unique contributions. First, through MPLens, we demonstrate how membership inference attack methods can be leveraged in adversarial machine learning. Second, through MPLens, we highlight…

Phase mapping for cardiac unipolar electrograms with neural network instead of phase transformation. (arXiv:1911.09731v1 [eess.SP])

Digital signal processing can be performed in two major ways. The first is a pipeline of signal transformations, and the second is machine learning approaches that require tagged data for training. This paper studies the third way. We generate a training dataset for a neural network in a series of numerical experiments and uses the…