2004.04165
Distributed peer review enhanced with natural language processing and machine learning
Kerzendorf, Patat, et al
While ancient scientists often had patrons to fund their work, peer review of proposals for the allocation of resources is a foundation of modern science. A very common method is that proposals are evaluated by a small panel of experts (due to logistics and funding limitations) nominated by the grant-giving institutions. The expert panel process introduces several issues - most notably: 1) biases introduced in the selection of the panel. 2) experts have to read a very large number of proposals. Distributed Peer Review promises to alleviate several of the described problems by distributing the task of reviewing among the proposers. Each proposer is given a limited number of proposals to review and rank. We present the result of an experiment running a machine-learning enhanced distributed peer review process for allocation of telescope time at the European Southern Observatory. In this work, we show that the distributed peer review is statistically the same as a `traditional' panel, that our machine learning algorithm can predict expertise of reviewers with a high success rate, and we find that seniority and reviewer expertise have an influence on review quality. The general experience has been overwhelmingly praised from the participating community (using an anonymous feedback mechanism).
Distributed peer review enhanced with natural language processing and machine learning
Kerzendorf, Patat, et al
While ancient scientists often had patrons to fund their work, peer review of proposals for the allocation of resources is a foundation of modern science. A very common method is that proposals are evaluated by a small panel of experts (due to logistics and funding limitations) nominated by the grant-giving institutions. The expert panel process introduces several issues - most notably: 1) biases introduced in the selection of the panel. 2) experts have to read a very large number of proposals. Distributed Peer Review promises to alleviate several of the described problems by distributing the task of reviewing among the proposers. Each proposer is given a limited number of proposals to review and rank. We present the result of an experiment running a machine-learning enhanced distributed peer review process for allocation of telescope time at the European Southern Observatory. In this work, we show that the distributed peer review is statistically the same as a `traditional' panel, that our machine learning algorithm can predict expertise of reviewers with a high success rate, and we find that seniority and reviewer expertise have an influence on review quality. The general experience has been overwhelmingly praised from the participating community (using an anonymous feedback mechanism).
2004.04253
Extragalactic cosmic rays diffusing from two populations of sources
Mollerach, Roulet
We consider the possibility of explaining the observed spectrum and composition of the cosmic rays with energies above $10^{17}$ eV in terms of two different extragalactic populations of sources in the presence of a turbulent intergalactic magnetic field (including also a fading Galactic cosmic-ray component). The populations are considered to be the superposition of different nuclear species having rigidity dependent spectra. The first extragalactic population is dominant in the energy range $10^{17}$ to $10^{18}$ eV and consists of sources having a relatively large density ($> 10^{-3}$ Mpc$^{-3}$) and a steep spectrum. The second extragalactic population dominates the cosmic ray flux above few EeV, it has a harder spectral slope and has a high-energy cutoff at few $Z$ EeV (where $eZ$ is the associated cosmic ray charge). This population has a lower density of sources ($<10^{-4}$ Mpc$^{-3}$), so that the typical intersource separation is larger than few tens of Mpc, being significantly affected by a magnetic horizon effect that strongly suppresses its flux for energies below $\sim Z$ EeV. We discuss how this scenario could be reconciled with the values of the cosmic-ray source spectral indices that are expected to result from the diffusive shock acceleration mechanism.
Extragalactic cosmic rays diffusing from two populations of sources
Mollerach, Roulet
We consider the possibility of explaining the observed spectrum and composition of the cosmic rays with energies above $10^{17}$ eV in terms of two different extragalactic populations of sources in the presence of a turbulent intergalactic magnetic field (including also a fading Galactic cosmic-ray component). The populations are considered to be the superposition of different nuclear species having rigidity dependent spectra. The first extragalactic population is dominant in the energy range $10^{17}$ to $10^{18}$ eV and consists of sources having a relatively large density ($> 10^{-3}$ Mpc$^{-3}$) and a steep spectrum. The second extragalactic population dominates the cosmic ray flux above few EeV, it has a harder spectral slope and has a high-energy cutoff at few $Z$ EeV (where $eZ$ is the associated cosmic ray charge). This population has a lower density of sources ($<10^{-4}$ Mpc$^{-3}$), so that the typical intersource separation is larger than few tens of Mpc, being significantly affected by a magnetic horizon effect that strongly suppresses its flux for energies below $\sim Z$ EeV. We discuss how this scenario could be reconciled with the values of the cosmic-ray source spectral indices that are expected to result from the diffusive shock acceleration mechanism.
2004.04615
Cross-matching of OGLE III and GAIA catalogues: investigation of dark-lens microlensing candidates
Dehghani, Rahvar
In this work, we use $13$ microlensing candidates with dark lenses from OGLE III catalogue \citep{2016MNRAS.458.3012W} and cross-match them with GAIA catalogue. We identify the microlensing source stars in GAIA catalogue by comparing the coordinate and the magnitude of stars and use the proper motion and the parallax parameters of the source stars. Combining with the microlensing light curves as well as microlensing parallax effect, we determine the mass and the distance of lenses from Earth. We conclude that the lens of some of microlensing events can be a blackhole.
2004.04702
Cosmology with the Wide-Field Infrared Survey Telescope -- Synergies with the Rubin Observatory Legacy Survey of Space and TIme
Eifler, et al
We explore synergies between the space-based Wide-Field Infrared Survey Telescope (WFIRST) and the ground-based Rubin Observatory Legacy Survey of Space and Time (LSST). In particular, we consider a scenario where the currently envisioned survey strategy for WFIRST's High Latitude Survey (HLS), i.e., 2000 square degrees in four narrow photometric bands is altered in favor of a strategy that combines rapid coverage of the LSST area (to full LSST depth) in one band. We find that a 5-month WFIRST survey in the W-band can cover the full LSST survey area providing high-resolution imaging for >95% of the LSST Year 10 gold galaxy sample. We explore a second, more ambitious scenario where WFIRST spends 1.5 years covering the LSST area. For this second scenario we quantify the constraining power on dark energy equation of state parameters from a joint weak lensing and galaxy clustering analysis, and compare it to an LSST-only survey and to the Reference WFIRST HLS survey. Our survey simulations are based on the WFIRST exposure time calculator and redshift distributions from the CANDELS catalog. Our statistical uncertainties account for higher-order correlations of the density field, and we include a wide range of systematic effects, such as uncertainties in shape and redshift measurements, and modeling uncertainties of astrophysical systematics, such as galaxy bias, intrinsic galaxy alignment, and baryonic physics. Assuming the 5-month WFIRST wide scenario, we find a significant increase in constraining power for the joint LSST+WFIRST wide survey compared to LSST Y10 (FoM(Wwide)= 2.4 FoM(LSST)) and compared to LSST+WFIRST HLS (FoM(Wwide})= 5.5 FoM(HLS)).
No comments:
Post a Comment