Source author record

Joseph Anderson

Joseph Anderson appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

astro-ph.HE astro-ph.SR Machine Learning astro-ph.CO Data Structures and Algorithms astro-ph.EP astro-ph.GA astro-ph.IM Computation math.ST physics.ins-det Statistics Theory

Catalog footprint

What is connected

7works

12topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2020arXiv

The Tidal Disruption Event AT 2018hyz II: Light Curve Modeling of a Partially Disrupted Star

AT 2018hyz (=ASASSN-18zj) is a tidal disruption event (TDE) located in the nucleus of a quiescent E+A galaxy at a redshift of $z = 0.04573$, first detected by the All-Sky Automated Survey for Supernovae (ASAS-SN). We present optical+UV photometry of the transient, as well as an X-ray spectrum and radio upper limits. The bolometric light curve of AT 2018hyz is comparable to other known TDEs and declines at a rate consistent with a $t^{-5/3}$ at early times, emitting a total radiated energy of $E = 9\times10^{50}$ erg. An excess bump appears in the UV light curve about 50 days after bolometric peak, followed by a flattening beyond 250 days. The light curve shows an excess bump in the UV about 50 days after bolometric peak lasting for at least 100 days, which may be related to an outflow. We detect a constant X-ray source present for at least 86 days. The X-ray spectrum shows a total unabsorbed flux of $\sim 4\times10^{-14}$ erg cm$^{-2}$ s$^{-1}$ and is best fit by a blackbody plus power-law model with a photon index of $Γ= 0.8$. A thermal X-ray model is unable to account for photons $> 1$ keV, while the radio non-detection favors inverse-Compton scattering rather than a jet for the non-thermal component. We model the optical and UV light curves using the Modular Open-Source Fitter for Transients (MOSFiT) and find a best fit for a black hole of $5.2\times10^6$ M$_\odot$ partially disrupting a $0.1$ M$_\odot$ star (stripping a mass of $\sim 0.01$ M$_\odot$ for the inferred impact parameter, $β=0.6$). The low optical depth implied by the small debris mass may explain how we are able to see hydrogen emission with disk-like line profiles in the spectra of AT 2018hyz (see our companion paper, Short et al.~2020).

preprint2016arXiv

Regaining the FORS: making optical ground-based transmission spectroscopy of exoplanets with VLT+FORS2 possible again

Transmission spectroscopy facilitates the detection of molecules and/or clouds in the atmospheres of exoplanets. Such studies rely heavily on space-based or large ground-based observatories, as one needs to perform time- resolved, high signal-to-noise spectroscopy. The FORS2 instrument at ESO's Very Large Telescope is the obvious choice for performing such studies, and was indeed pioneering the field in 2010. After that, however, it was shown to suffer from systematic errors caused by the Longitudinal Atmospheric Dispersion Corrector (LADC). This was successfully addressed, leading to a renewed interest for this instrument as shown by the number of proposals submitted to perform transmission spectroscopy of exoplanets. We present here the context, the problem and how we solved it, as well as the recent results obtained. We finish by providing tips for an optimum strategy to do transmission spectroscopy with FORS2, in the hope that FORS2 may become the instrument of choice for ground-based transmission spectroscopy of exoplanets.

preprint2015arXiv

Heavy-tailed Independent Component Analysis

Independent component analysis (ICA) is the problem of efficiently recovering a matrix $A \in \mathbb{R}^{n\times n}$ from i.i.d. observations of $X=AS$ where $S \in \mathbb{R}^n$ is a random vector with mutually independent coordinates. This problem has been intensively studied, but all existing efficient algorithms with provable guarantees require that the coordinates $S_i$ have finite fourth moments. We consider the heavy-tailed ICA problem where we do not make this assumption, about the second moment. This problem also has received considerable attention in the applied literature. In the present work, we first give a provably efficient algorithm that works under the assumption that for constant $γ> 0$, each $S_i$ has finite $(1+γ)$-moment, thus substantially weakening the moment requirement condition for the ICA problem to be solvable. We then give an algorithm that works under the assumption that matrix $A$ has orthogonal columns but requires no moment assumptions. Our techniques draw ideas from convex geometry and exploit standard properties of the multivariate spherical Gaussian distribution in a novel way.

preprint2014arXiv

Environments of interacting transients: Impostors and type IIn supernovae

This paper presents one of the first environmental analyses of the locations of the class of `interacting transients', namely type IIn supernovae and supernova Impostors. We discuss the association of these transients with star formation, host galaxy type, metallicity, and the locations of each event within the respective host. Given the frequent assumption of very high mass progenitors for these explosions from various studies, most notably a direct progenitor detection, it is interesting to note the weak association of these subtypes with star formation as traced by Hα emission, particularly in comparison with type Ic supernovae, which trace the Hα emission and are thought to arise from high mass progenitors. The radial distributions of these transients compared to type Ic supernovae are also very different. This provides evidence for the growing hypothesis that these `interacting transients' are in fact comprised of a variety of progenitor systems. The events contained within this sample are discussed in detail, where information in the literature exists, and compared to the environmental data provided. Impostors are found to split into two main classes, in terms of environment: SN2008S-like Impostors fall on regions of zero Hα emission, whereas η-Carina-like Impostors all fall on regions with positive Hα emission. We also find indications that the Impostor class originate from lower metallicity environments than type IIn, Ic and IIP SNe.

preprint2014arXiv

The More, the Merrier: the Blessing of Dimensionality for Learning Large Gaussian Mixtures

In this paper we show that very large mixtures of Gaussians are efficiently learnable in high dimension. More precisely, we prove that a mixture with known identical covariance matrices whose number of components is a polynomial of any fixed degree in the dimension n is polynomially learnable as long as a certain non-degeneracy condition on the means is satisfied. It turns out that this condition is generic in the sense of smoothed complexity, as soon as the dimensionality of the space is high enough. Moreover, we prove that no such condition can possibly exist in low dimension and the problem of learning the parameters is generically hard. In contrast, much of the existing work on Gaussian Mixtures relies on low-dimensional projections and thus hits an artificial barrier. Our main result on mixture recovery relies on a new "Poissonization"-based technique, which transforms a mixture of Gaussians to a linear map of a product distribution. The problem of learning this map can be efficiently solved using some recent results on tensor decompositions and Independent Component Analysis (ICA), thus giving an algorithm for recovering the mixture. In addition, we combine our low-dimensional hardness results for Gaussian mixtures with Poissonization to show how to embed difficult instances of low-dimensional Gaussian mixtures into the ICA setting, thus establishing exponential information-theoretic lower bounds for underdetermined ICA in low dimension. To the best of our knowledge, this is the first such result in the literature. In addition to contributing to the problem of Gaussian mixture learning, we believe that this work is among the first steps toward better understanding the rare phenomenon of the "blessing of dimensionality" in the computational aspects of statistical inference.

preprint2013arXiv

Efficient learning of simplices

We show an efficient algorithm for the following problem: Given uniformly random points from an arbitrary n-dimensional simplex, estimate the simplex. The size of the sample and the number of arithmetic operations of our algorithm are polynomial in n. This answers a question of Frieze, Jerrum and Kannan [FJK]. Our result can also be interpreted as efficiently learning the intersection of n+1 half-spaces in R^n in the model where the intersection is bounded and we are given polynomially many uniform samples from it. Our proof uses the local search technique from Independent Component Analysis (ICA), also used by [FJK]. Unlike these previous algorithms, which were based on analyzing the fourth moment, ours is based on the third moment. We also show a direct connection between the problem of learning a simplex and ICA: a simple randomized reduction to ICA from the problem of learning a simplex. The connection is based on a known representation of the uniform measure on a simplex. Similar representations lead to a reduction from the problem of learning an affine transformation of an n-dimensional l_p ball to ICA.

preprint2012arXiv

Evidence for asymmetric distribution of circumstellar material around Type Ia Supernovae

We study the properties of low-velocity material in the line of sight towards nearby Type Ia Supernovae (SNe Ia) that have measured late phase nebular velocity shifts (v_neb), thought to be an environment-independent observable. We have found that the distribution of equivalent widths of narrow blended Na I D1 & D2 and Ca II H & K absorption lines differs significantly between those SNe Ia with negative and positive v_neb, with generally stronger absorption for SNe Ia with v_neb > 0. A similar result had been found previously for the distribution of colors of SNe Ia, which was interpreted as a dependence of the temperature of the ejecta with viewing angle. Our work suggests that: 1) a significant part of these differences in color should be attributed to extinction, 2) this extinction is caused by an asymmetric distribution of circumstellar material (CSM) and 3) the CSM absorption is generally stronger on the side of the ejecta opposite to where the ignition occurs. Since it is difficult to explain 3) via any known physical processes that occur before explosion, we argue that the asymmetry of the CSM is originated after explosion by a stronger ionizing flux on the side of the ejecta where ignition occurs, probably due to a stronger shock breakout and/or more exposed radioactive material on one side of the ejecta. This result has important implications for both progenitor and explosion models.