-
An exploration of the properties of cluster profiles for the thermal and kinetic Sunyaev-Zel'dovich effects
Authors:
Billy K. K. Lee,
William R. Coulton,
Leander Thiele,
Shirley Ho
Abstract:
With the advent of high-resolution, low-noise CMB measurements, the ability to extract cosmological information from thermal Sunyaev-Zel'dovich effect and kinetic Sunyaev-Zel'dovich effect will be limited not by statistical uncertainties but rather by systematic and theoretical uncertainties. The theoretical uncertainty is driven by the lack of knowledge about the electron pressure and density. Th…
▽ More
With the advent of high-resolution, low-noise CMB measurements, the ability to extract cosmological information from thermal Sunyaev-Zel'dovich effect and kinetic Sunyaev-Zel'dovich effect will be limited not by statistical uncertainties but rather by systematic and theoretical uncertainties. The theoretical uncertainty is driven by the lack of knowledge about the electron pressure and density. Thus we explore the electron pressure and density distributions in the IllustrisTNG hydrodynamical simulations, and we demonstrate that the cluster properties exhibit a strong dependence on the halo concentration -- providing some of the first evidence of cluster assembly bias in the electron pressure and density. Further, our work shows evidence for a broken power-law mass dependence, with lower pressure in lower mass halos than previous work and a strong evolution with mass of the radial correlations in the electron density and pressure. Both of these effects highlight the differing impact of active galactic nuclei and supernova feedback on the gas in galaxy groups compared to massive clusters. We verified that we see qualitatively similar features in the SIMBA hydro-dynamical simulations, suggesting these effects could be generic features. Finally, we provide a parametric formula for the electron pressure and density profile as a function of dark matter halo mass, halo concentration, and redshift. These fitting formulae can reproduce the distribution of density and pressure of clusters and will be useful in extracting cosmological information from upcoming CMB surveys.
△ Less
Submitted 3 May, 2022;
originally announced May 2022.
-
Can the GW190814 secondary component be a bosonic dark matter admixed compact star?
Authors:
Billy K. K. Lee,
Ming-chung Chu,
Lap-Ming Lin
Abstract:
We investigate whether the recently observed 2.6 $M_\odot$ compact object in the gravitational-wave event GW190814 can be a bosonic dark matter admixed compact star. By considering the three constraints in mass, radius and stability of such an object, we find that if the dark matter is made of QCD axions, their particle mass $m$ is constrained to a range that has already been ruled out by the inde…
▽ More
We investigate whether the recently observed 2.6 $M_\odot$ compact object in the gravitational-wave event GW190814 can be a bosonic dark matter admixed compact star. By considering the three constraints in mass, radius and stability of such an object, we find that if the dark matter is made of QCD axions, their particle mass $m$ is constrained to a range that has already been ruled out by the independent constraint imposed by the stellar-mass black hole superradiance process. The 2.6 $M_\odot$ object can still be a neutron star admixed with at least 2.0 $M_\odot$ of dark matter made of axion-like particles (or even a pure axion-like particle star) if $2 \times 10^{-11}$ eV $\leq m \leq 2.4 \times 10^{-11}$ eV ($2.9 \times 10^{-11}$ eV $\leq m \leq 3.2 \times 10^{-11}$ eV) and with decay constant $f \geq 8 \times 10^{17}$ GeV.
△ Less
Submitted 11 October, 2021;
originally announced October 2021.
-
Machine Learning for Scent: Learning Generalizable Perceptual Representations of Small Molecules
Authors:
Benjamin Sanchez-Lengeling,
Jennifer N. Wei,
Brian K. Lee,
Richard C. Gerkin,
Alán Aspuru-Guzik,
Alexander B. Wiltschko
Abstract:
Predicting the relationship between a molecule's structure and its odor remains a difficult, decades-old task. This problem, termed quantitative structure-odor relationship (QSOR) modeling, is an important challenge in chemistry, impacting human nutrition, manufacture of synthetic fragrance, the environment, and sensory neuroscience. We propose the use of graph neural networks for QSOR, and show t…
▽ More
Predicting the relationship between a molecule's structure and its odor remains a difficult, decades-old task. This problem, termed quantitative structure-odor relationship (QSOR) modeling, is an important challenge in chemistry, impacting human nutrition, manufacture of synthetic fragrance, the environment, and sensory neuroscience. We propose the use of graph neural networks for QSOR, and show they significantly out-perform prior methods on a novel data set labeled by olfactory experts. Additional analysis shows that the learned embeddings from graph neural networks capture a meaningful odor space representation of the underlying relationship between structure and odor, as demonstrated by strong performance on two challenging transfer learning tasks. Machine learning has already had a large impact on the senses of sight and sound. Based on these early results with graph neural networks for molecular properties, we hope machine learning can eventually do for olfaction what it has already done for vision and hearing.
△ Less
Submitted 25 October, 2019; v1 submitted 23 October, 2019;
originally announced October 2019.
-
A quasi-probability for the arrival time problem with links to backflow and the Leggett-Garg inequalities
Authors:
J. J. Halliwell,
H. Beck,
B. K. B. Lee,
S. O'Brien
Abstract:
The arrival time problem for the free particle in one dimension may be formulated as the problem of determining a joint probability for the particle being found on opposite sides of the $x$-axis at two different times. We explore this problem using a two-time quasi-probability linear in the projection operators, a natural counterpart of the corresponding classical problem. We show that it can be m…
▽ More
The arrival time problem for the free particle in one dimension may be formulated as the problem of determining a joint probability for the particle being found on opposite sides of the $x$-axis at two different times. We explore this problem using a two-time quasi-probability linear in the projection operators, a natural counterpart of the corresponding classical problem. We show that it can be measured either indirectly, by measuring its moments in different experiments, or directly, in a single experiment using a pair of sequential measurements in which the first measurement is weak (or more generally, ambiguous). We argue that when positive, it corresponds to a measurement-independent arrival time probability. For small time intervals it coincides approximately with the time-averaged current, in agreement with semiclassical expectations. The quasi-probability can be negative and we exhibit a number of situations in which this is the case. We interpret these situations as the presence of `quantumness', in which the arrival time probability is not properly defined in a measurement-independent manner. Backflow states, in which the current flows in the direction opposite to the momentum, are shown to provide an interesting class of examples such situations. We also show that the quasi-probability is closely linked to a set of two-time Leggett-Garg inequalities, which test for macroscopic realism.
△ Less
Submitted 25 October, 2018;
originally announced October 2018.
-
AutoGraph: Imperative-style Coding with Graph-based Performance
Authors:
Dan Moldovan,
James M Decker,
Fei Wang,
Andrew A Johnson,
Brian K Lee,
Zachary Nado,
D Sculley,
Tiark Rompf,
Alexander B Wiltschko
Abstract:
There is a perceived trade-off between machine learning code that is easy to write, and machine learning code that is scalable or fast to execute. In machine learning, imperative style libraries like Autograd and PyTorch are easy to write, but suffer from high interpretive overhead and are not easily deployable in production or mobile settings. Graph-based libraries like TensorFlow and Theano bene…
▽ More
There is a perceived trade-off between machine learning code that is easy to write, and machine learning code that is scalable or fast to execute. In machine learning, imperative style libraries like Autograd and PyTorch are easy to write, but suffer from high interpretive overhead and are not easily deployable in production or mobile settings. Graph-based libraries like TensorFlow and Theano benefit from whole-program optimization and can be deployed broadly, but make expressing complex models more cumbersome. We describe how the use of staged programming in Python, via source code transformation, offers a midpoint between these two library design patterns, capturing the benefits of both. A key insight is to delay all type-dependent decisions until runtime, via dynamic dispatch. We instantiate these principles in AutoGraph, a software system that improves the programming experience of the TensorFlow library, and demonstrate usability improvements with no loss in performance compared to native TensorFlow graphs. We also show that our system is backend agnostic, and demonstrate targeting an alternate IR with characteristics not found in TensorFlow graphs.
△ Less
Submitted 26 March, 2019; v1 submitted 16 October, 2018;
originally announced October 2018.
