-
Lessons Learned in Performing a Trustworthy AI and Fundamental Rights Assessment
Authors:
Marjolein Boonstra,
Frédérick Bruneault,
Subrata Chakraborty,
Tjitske Faber,
Alessio Gallucci,
Eleanore Hickman,
Gerard Kema,
Hee** Kim,
Jaap Kooiker,
Elisabeth Hildt,
Annegret Lamadé,
Emilie Wiinblad Mathez,
Florian Möslein,
Genien Pathuis,
Giovanni Sartor,
Marijke Steege,
Alice Stocco,
Willy Tadema,
Jarno Tuimala,
Isabel van Vledder,
Dennis Vetter,
Jana Vetter,
Magnus Westerlund,
Roberto V. Zicari
Abstract:
This report shares the experiences, results and lessons learned in conducting a pilot project ``Responsible use of AI'' in cooperation with the Province of Friesland, Rijks ICT Gilde-part of the Ministry of the Interior and Kingdom Relations (BZK) (both in The Netherlands) and a group of members of the Z-Inspection$^{\small{\circledR}}$ Initiative. The pilot project took place from May 2022 throug…
▽ More
This report shares the experiences, results and lessons learned in conducting a pilot project ``Responsible use of AI'' in cooperation with the Province of Friesland, Rijks ICT Gilde-part of the Ministry of the Interior and Kingdom Relations (BZK) (both in The Netherlands) and a group of members of the Z-Inspection$^{\small{\circledR}}$ Initiative. The pilot project took place from May 2022 through January 2023. During the pilot, the practical application of a deep learning algorithm from the province of Frŷslan was assessed. The AI maps heathland grassland by means of satellite images for monitoring nature reserves. Environmental monitoring is one of the crucial activities carried on by society for several purposes ranging from maintaining standards on drinkable water to quantifying the CO2 emissions of a particular state or region. Using satellite imagery and machine learning to support decisions is becoming an important part of environmental monitoring. The main focus of this report is to share the experiences, results and lessons learned from performing both a Trustworthy AI assessment using the Z-Inspection$^{\small{\circledR}}$ process and the EU framework for Trustworthy AI, and combining it with a Fundamental Rights assessment using the Fundamental Rights and Algorithms Impact Assessment (FRAIA) as recommended by the Dutch government for the use of AI algorithms by the Dutch public authorities.
△ Less
Submitted 22 April, 2024;
originally announced April 2024.
-
The Size and Shape of the Milky Way Disk and Halo from M-type Brown Dwarfs in the BoRG Survey
Authors:
Isabel van Vledder,
Dieuwertje van der Vlugt,
B. W. Holwerda,
M. A. Kenworthy,
R. J. Bouwens,
M. Trenti
Abstract:
We have identified 274 M-type Brown Dwarfs in the Hubble Space Telescope's Wide Field Camera 3 (WFC3) pure parallel fields from the Brightest of Reionizing Galaxies (BoRG) survey for high redshift galaxies. These are near-infrared observations with multiple lines-of-sight out of our Milky Way. Using these observed M-type Brown Dwarfs we fitted a Galactic disk and halo model with a Markov chain Mon…
▽ More
We have identified 274 M-type Brown Dwarfs in the Hubble Space Telescope's Wide Field Camera 3 (WFC3) pure parallel fields from the Brightest of Reionizing Galaxies (BoRG) survey for high redshift galaxies. These are near-infrared observations with multiple lines-of-sight out of our Milky Way. Using these observed M-type Brown Dwarfs we fitted a Galactic disk and halo model with a Markov chain Monte Carlo (MCMC) analysis. This model worked best with the scale length of the disk fixed at $h$ = 2.6 kpc. For the scale height of the disk, we found $z_0 = 0.29^{+0.02}_{-0.019}$ kpc and for the central number density $ρ_0 = 0.29^{+0.20}_{-0.13}$ \#/pc$^3$. For the halo we derived a flattening parameter $κ$ = 0.45$\pm{0.04}$ and a power-law index $p$ = 2.4$\pm{0.07}$. We found the fraction of M-type brown dwarfs in the local density that belong to the halo to be $f_{h}$ = 0.0075$^{+0.0025}_{-0.0019}$. We found no correlation between subtype of M-dwarf and any model parameters.
The total number of M-type Brown Dwarfs in the disk and halo was determined to be $58.2^{+9.81}_{-6.70} \times10^{9}$. We found an upper limit for the fraction of M-type Brown Dwarfs in the halo of 7$^{+5}_{-4}$\%. The upper limit for the total Galactic Disk mass in M-dwarfs is $4.34^{+0.73}_{-0.5}\times10^{9}$ $M_{\odot}$, assuming all M-type Brown Dwarfs have a mass of $80 M_J$.
△ Less
Submitted 16 March, 2016;
originally announced March 2016.
-
Milky Way Red Dwarfs in the BoRG Survey; Galactic scale-height and the distribution of dwarfs stars in WFC3 imaging
Authors:
B. W. Holwerda,
M. Trenti,
W. Clarkson,
K. Sahu,
L. Bradley,
M. Stiavelli,
N. Pirzkal,
G. De Marchi,
M. Andersen,
R. Bouwens,
R. Ryan,
I. van Vledder,
D. van der Vlugt
Abstract:
We present a tally of Milky Way late-type dwarf stars in 68 WFC3 pure-parallel fields (227 arcmin^2) from the Brightest of Reionizing Galaxies (BoRG) survey for high-redshift galaxies. Using spectroscopically identified M-dwarfs in two public surveys, the CANDELS and the ERS mosaics, we identify a morphological selection criterion using the half-light radius (r50), a near-infrared J-H, G-J color r…
▽ More
We present a tally of Milky Way late-type dwarf stars in 68 WFC3 pure-parallel fields (227 arcmin^2) from the Brightest of Reionizing Galaxies (BoRG) survey for high-redshift galaxies. Using spectroscopically identified M-dwarfs in two public surveys, the CANDELS and the ERS mosaics, we identify a morphological selection criterion using the half-light radius (r50), a near-infrared J-H, G-J color region where M-dwarfs are found, and a V-J relation with M-dwarf subtype. We apply this morphological selection of stellar objects, color-color selection of M-dwarfs and optical-near-infrared color subty** to compile a catalog of 274 M-dwarfs belonging to the disk of the Milky Way with a limiting magnitude of m_F125W < 24.
Based on the M-dwarfs statistics, we conclude that (a) the previously identified North/South discrepancy in M-dwarf numbers persists in our sample; there are more M-dwarfs in the Northern fields on average than in Southern ones, (b) the Milky Way's single disk scale-height for M-dwarfs is 0.3-4 kpc, depending on sub-type, (c) {\bf ERRATUM:} we present corrected coordinates (AstroPy) and distances and find a constant $z_0$=600 pc for all types. (d) a second component is visible in the vertical distribution, with a different, much higher scale-height. We report the M-dwarf component of the Sagittarius stream in one of our fields with 11 confirmed M-dwarfs, 7 of which are at the stream's distance.
The dwarf scale-height and the relative low incidence in our fields of L- and T-dwarfs in these fields makes it unlikely that these stars will be interlopers in great numbers in color-selected samples of high-redshift galaxies. The relative ubiquity of M-dwarfs however will make them ideal tracers of Galactic Halo substructure with EUCLID and reference stars for JWST observations.
△ Less
Submitted 8 February, 2016; v1 submitted 17 February, 2014;
originally announced February 2014.
