-
Unpacking Human-AI Interaction in Safety-Critical Industries: A Systematic Literature Review
Authors:
Tita A. Bach,
Jenny K. Kristiansen,
Aleksandar Babic,
Alon Jacovi
Abstract:
Ensuring quality human-AI interaction (HAII) in safety-critical industries is essential. Failure to do so can lead to catastrophic and deadly consequences. Despite this urgency, what little research there is on HAII is fragmented and inconsistent. We present here a survey of that literature and recommendations for research best practices that will improve the field. We divided our investigation in…
▽ More
Ensuring quality human-AI interaction (HAII) in safety-critical industries is essential. Failure to do so can lead to catastrophic and deadly consequences. Despite this urgency, what little research there is on HAII is fragmented and inconsistent. We present here a survey of that literature and recommendations for research best practices that will improve the field. We divided our investigation into the following research areas: (1) terms used to describe HAII, (2) primary roles of AI-enabled systems, (3) factors that influence HAII, and (4) how HAII is measured. Additionally, we described the capabilities and maturity of the AI-enabled systems used in safety-critical industries discussed in these articles. We found that no single term is used across the literature to describe HAII and some terms have multiple meanings. According to our literature, five factors influence HAII: user characteristics and background (e.g., user personality, perceptions), AI interface and features (e.g., interactive UI design), AI output (e.g., accuracy, actionable recommendations), explainability and interpretability (e.g., level of detail, user understanding), and usage of AI (e.g., heterogeneity of environments and user needs). HAII is most commonly measured with user-related subjective metrics (e.g., user perception, trust, and attitudes), and AI-assisted decision-making is the most common primary role of AI-enabled systems. Based on this review, we conclude that there are substantial research gaps in HAII. Researchers and developers need to codify HAII terminology, involve users throughout the AI lifecycle (especially during development), and tailor HAII in safety-critical industries to the users and environments.
△ Less
Submitted 5 October, 2023;
originally announced October 2023.
-
High-performance computing for super-resolution microscopy on a cluster of computers
Authors:
Quan Do,
Jon Ivar Kristiansen,
Krishna Agarwal,
Phuong Hoai Ha
Abstract:
Multiple signal classification algorithm (MUSICAL) provides a super-resolution microscopy method. In the previous research, MUSICAL has enabled data-parallelism well on a desktop computer or a Linux-based server. However, the running time needs to be shorter. This paper will develop a new parallel MUSICAL with high efficiency and scalability on a cluster of computers. We achieve the purpose by usi…
▽ More
Multiple signal classification algorithm (MUSICAL) provides a super-resolution microscopy method. In the previous research, MUSICAL has enabled data-parallelism well on a desktop computer or a Linux-based server. However, the running time needs to be shorter. This paper will develop a new parallel MUSICAL with high efficiency and scalability on a cluster of computers. We achieve the purpose by using the optimal speed of the cluster cores, the latest parallel programming techniques, and the high-performance computing libraries, such as the Intel Threading Building Blocks (TBB), the Intel Math Kernel Library (MKL), and the unified parallel C++ (UPC++) for the cluster of computers. Our experimental results show that the new parallel MUSICAL achieves a speed-up of 240.29x within 10 seconds on the 256-core cluster with an efficiency of 93.86%. Our MUSICAL offers a high possibility for real-life applications to make super-resolution microscopy within seconds.
△ Less
Submitted 13 June, 2022; v1 submitted 7 June, 2022;
originally announced June 2022.
-
Cosmology with sterile neutrino masses from oscillation experiments
Authors:
Jostein R. Kristiansen,
Øystein Elgarøy,
Carlo Giunti,
Marco Laveder
Abstract:
There are hints, both from cosmology and from neutrino oscillation experiments, that one or two sterile neutrinos with eV masses are favored. We consider the implications of combining data from short baseline neutrino experiments with cosmological observations. In particular we consider cosmological models where the equation of state of dark energy and the curvature are allowed to vary. We find th…
▽ More
There are hints, both from cosmology and from neutrino oscillation experiments, that one or two sterile neutrinos with eV masses are favored. We consider the implications of combining data from short baseline neutrino experiments with cosmological observations. In particular we consider cosmological models where the equation of state of dark energy and the curvature are allowed to vary. We find that the consistency of results from short baseline experiments and cosmology is dependent on the cosmological data sets that are used. When the two are consistent, we find that the inclusion of information from short baseline experiments only mildly affects the preferred values of the cosmological model parameters. A universe with a cosmological constant type dark energy and zero curvature is still fully consistent with the data we use. We also report that a thermalized sterile neutrino in the 10 eV range is fully consistent with a wide range of cosmological data as long as information on the matter power spectrum in not taken into account.
△ Less
Submitted 19 March, 2013;
originally announced March 2013.
-
Reactor sterile neutrinos, dark energy and the age of the universe
Authors:
Jostein R. Kristiansen,
Oystein Elgaroy
Abstract:
There are indications that the neutrino oscillation data from reactor experiments and the LSND and MiniBooNE experiments show a preference for two sterile neutrino species, both with masses in the eV region. We show that this result has a significant impact on some important cosmological parameters. Specifically, we use a combination of CMB, LSS and SN1A data and show that the existence of two lig…
▽ More
There are indications that the neutrino oscillation data from reactor experiments and the LSND and MiniBooNE experiments show a preference for two sterile neutrino species, both with masses in the eV region. We show that this result has a significant impact on some important cosmological parameters. Specifically, we use a combination of CMB, LSS and SN1A data and show that the existence of two light, sterile neutrinos would rule out the cosmological constant as dark energy at 95% confidence level, and lower the expansion age of the universe to 12.58 \pm 0.26 Gyr.
△ Less
Submitted 11 May, 2011; v1 submitted 4 April, 2011;
originally announced April 2011.
-
Do WMAP5 data favor neutrino mass and a coupling between Cold Dark Matter and Dark Energy?
Authors:
S. A. Bonometto,
G. La Vacca,
J. R. Kristiansen,
R. Mainini,
L. P. L. Colombo
Abstract:
We fit WMAP5 and related data by allowing for a CDM--DE coupling and non--zero neutrino masses, simultaneously. We find a significant correlation between these parameters, so that simultaneous higher coupling and ν--masses are allowed. Furthermore, models with a significant coupling and ν--mass are statistically favoured in respect to a cosmology with no coupling and negligible neutrino mass (ou…
▽ More
We fit WMAP5 and related data by allowing for a CDM--DE coupling and non--zero neutrino masses, simultaneously. We find a significant correlation between these parameters, so that simultaneous higher coupling and ν--masses are allowed. Furthermore, models with a significant coupling and ν--mass are statistically favoured in respect to a cosmology with no coupling and negligible neutrino mass (our best fits are: C ~ 1/2m_p, m_ν~ 0.12eV per flavor). We use a standard Monte Carlo Markov Chain approach, by assuming DE to be a scalar field self--interacting through Ratra--Peebles or SUGRA potentials.
△ Less
Submitted 18 November, 2009;
originally announced November 2009.
-
Dynamical Dark Energy model parameters with or without massive neutrinos
Authors:
G. La Vacca,
J. R. Kristiansen
Abstract:
We use WMAP5 and other cosmological data to constrain model parameters in quintessence cosmologies, focusing also on their shift when we allow for non-vanishing neutrino masses. The Ratra-Peebles (RP) and SUGRA potentials are used here, as examples of slowly or fastly varying state parameter w(a). Both potentials depend on an energy scale Λ. Here we confirm the results of previous analysis with…
▽ More
We use WMAP5 and other cosmological data to constrain model parameters in quintessence cosmologies, focusing also on their shift when we allow for non-vanishing neutrino masses. The Ratra-Peebles (RP) and SUGRA potentials are used here, as examples of slowly or fastly varying state parameter w(a). Both potentials depend on an energy scale Λ. Here we confirm the results of previous analysis with WMAP3 data on the upper limits on Λ, which turn out to be rather small (down to ~10^{-9} in RP cosmologies and ~10^{-5} for SUGRA). Our constraints on Λare not heavily affected by the inclusion of neutrino mass as a free parameter. On the contrary, when the neutrino mass degree of freedom is opened, significant shifts in the best-fit values of other parameters occur.
△ Less
Submitted 24 June, 2009;
originally announced June 2009.
-
Do data favor neutrino mass and a coupling between Cold Dark Matter and Dark Energy?
Authors:
G. La Vacca,
J. R. Kristiansen,
L. P. L. Colombo,
R. Mainini,
S. A. Bonometto
Abstract:
We allow simultaneously for a CDM--DE coupling and non--zero neutrino masses and find that significant coupling and neutrino mass are (slightly) statistically favoured in respect to a cosmology with no coupling and negligible neutrino mass (our best fits are: C~1/2m_p, m_ν~0.12eV each flavor). We assume DE to be a self--interacting scalar field and use a standard Monte Carlo Markov Chain approac…
▽ More
We allow simultaneously for a CDM--DE coupling and non--zero neutrino masses and find that significant coupling and neutrino mass are (slightly) statistically favoured in respect to a cosmology with no coupling and negligible neutrino mass (our best fits are: C~1/2m_p, m_ν~0.12eV each flavor). We assume DE to be a self--interacting scalar field and use a standard Monte Carlo Markov Chain approach.
△ Less
Submitted 18 June, 2009;
originally announced June 2009.
-
Coupling between cold dark matter and dark energy from neutrino mass experiments
Authors:
J. R. Kristiansen,
G. La Vacca,
L. P. L. Colombo,
R. Mainini,
S. A. Bonometto
Abstract:
We consider cosmological models with dynamical dark energy (dDE) coupled to cold dark matter (CDM), while simultaneously allowing neutrinos to be massive. Using a MCMC approach, we compare these models with a wide range of cosmological data sets. We find a strong correlation between this coupling strength and the neutrino mass. This correlation persists when BAO data are included in the analysis…
▽ More
We consider cosmological models with dynamical dark energy (dDE) coupled to cold dark matter (CDM), while simultaneously allowing neutrinos to be massive. Using a MCMC approach, we compare these models with a wide range of cosmological data sets. We find a strong correlation between this coupling strength and the neutrino mass. This correlation persists when BAO data are included in the analysis. We add then priors on $ν$ mass from particle experiments. The claimed detection of $ν$ mass from the Heidelberg-Moscow neutrinoless double--$β$ decay experiment would imply a 7--$8 σ$ detection of CDM-DE coupling. Similarly, the detection of $ν$ mass from coming KATRIN tritium $β$ decay experiment will imply a safe detection of a coupling in the dark sector. Previous attempts to accommodate cosmic phenomenology with such possible $ν$ mass data made recourse to a $w < -1$ eoS. We compare such an option with the coupling option and find that the latter allows a drastic improvement.
△ Less
Submitted 16 February, 2010; v1 submitted 16 February, 2009;
originally announced February 2009.
-
Do WMAP data favor neutrino mass and a coupling between Cold Dark Matter and Dark Energy?
Authors:
G. La Vacca,
J. R. Kristiansen,
L. P. L. Colombo,
R. Mainini,
S. A. Bonometto
Abstract:
Within the frame of cosmologies where Dark Energy (DE) is a self--interacting scalar field, we allow for a CDM--DE coupling and non--zero neutrino masses, simultaneously. In their 0--0 version, i.e. in the absence of coupling and neutrino mass, these cosmologies provide an excellent fit to WMAP, SNIa and deep galaxy sample spectra, at least as good as \LambdaCDM. When the new degrees of freedom…
▽ More
Within the frame of cosmologies where Dark Energy (DE) is a self--interacting scalar field, we allow for a CDM--DE coupling and non--zero neutrino masses, simultaneously. In their 0--0 version, i.e. in the absence of coupling and neutrino mass, these cosmologies provide an excellent fit to WMAP, SNIa and deep galaxy sample spectra, at least as good as \LambdaCDM. When the new degrees of freedom are open, we find that CDM--DE coupling and significant neutrino masses (~0.1eV per νspecies) are at least as likely as the 0--0 option and, in some cases, even statistically favoured. Results are obtained by using a Monte Carlo Markov Chain approach.
△ Less
Submitted 16 February, 2009;
originally announced February 2009.
-
CMBPol Mission Concept Study: Prospects for polarized foreground removal
Authors:
J. Dunkley,
A. Amblard,
C. Baccigalupi,
M. Betoule,
D. Chuss,
A. Cooray,
J. Delabrouille,
C. Dickinson,
G. Dobler,
J. Dotson,
H. K. Eriksen,
D. Finkbeiner,
D. Fixsen,
P. Fosalba,
A. Fraisse,
C. Hirata,
A. Kogut,
J. Kristiansen,
C. Lawrence,
A. M. Magalhaes,
M. A. Miville-Deschenes,
S. Meyer,
A. Miller,
S. K. Naess,
L. Page
, et al. (6 additional authors not shown)
Abstract:
In this report we discuss the impact of polarized foregrounds on a future CMBPol satellite mission. We review our current knowledge of Galactic polarized emission at microwave frequencies, including synchrotron and thermal dust emission. We use existing data and our understanding of the physical behavior of the sources of foreground emission to generate sky templates, and start to assess how wel…
▽ More
In this report we discuss the impact of polarized foregrounds on a future CMBPol satellite mission. We review our current knowledge of Galactic polarized emission at microwave frequencies, including synchrotron and thermal dust emission. We use existing data and our understanding of the physical behavior of the sources of foreground emission to generate sky templates, and start to assess how well primordial gravitational wave signals can be separated from foreground contaminants for a CMBPol mission. At the estimated foreground minimum of ~100 GHz, the polarized foregrounds are expected to be lower than a primordial polarization signal with tensor-to-scalar ratio r=0.01, in a small patch (~1%) of the sky known to have low Galactic emission. Over 75% of the sky we expect the foreground amplitude to exceed the primordial signal by about a factor of eight at the foreground minimum and on scales of two degrees. Only on the largest scales does the polarized foreground amplitude exceed the primordial signal by a larger factor of about 20. The prospects for detecting an r=0.01 signal including degree-scale measurements appear promising, with 5 sigma_r ~0.003 forecast from multiple methods. A mission that observes a range of scales offers better prospects from the foregrounds perspective than one targeting only the lowest few multipoles. We begin to explore how optimizing the composition of frequency channels in the focal plane can maximize our ability to perform component separation, with a range of typically 40 < nu < 300 GHz preferred for ten channels. Foreground cleaning methods are already in place to tackle a CMBPol mission data set, and further investigation of the optimization and detectability of the primordial signal will be useful for mission design.
△ Less
Submitted 2 November, 2009; v1 submitted 24 November, 2008;
originally announced November 2008.
-
Velocity dispersions in a cluster of stars: How fast could Usain Bolt have run?
Authors:
H. K. Eriksen,
J. R. Kristiansen,
O. Langangen,
I. K. Wehus
Abstract:
Since that very memorable day at the Bei**g 2008 Olympics, a big question on every sports commentator's mind has been "What would the 100 meter dash world record have been, had Usain Bolt not celebrated at the end of his race?" Glen Mills, Bolt's coach suggested at a recent press conference that the time could have been 9.52 seconds or better. We revisit this question by measuring Bolt's positi…
▽ More
Since that very memorable day at the Bei**g 2008 Olympics, a big question on every sports commentator's mind has been "What would the 100 meter dash world record have been, had Usain Bolt not celebrated at the end of his race?" Glen Mills, Bolt's coach suggested at a recent press conference that the time could have been 9.52 seconds or better. We revisit this question by measuring Bolt's position as a function of time using footage of the run, and then extrapolate into the last two seconds based on two different assumptions. First, we conservatively assume that Bolt could have maintained Richard Thompson's, the runner-up, acceleration during the end of the race. Second, based on the race development prior to the celebration, we assume that he could also have kept an acceleration of 0.5 m/s^2 higher than Thompson. In these two cases, we find that the new world record would have been 9.61 +/- 0.04 and 9.55 +/- 0.04 seconds, respectively, where the uncertainties denote 95% statistical errors.
△ Less
Submitted 2 September, 2008; v1 submitted 1 September, 2008;
originally announced September 2008.
-
Constraining primordial magnetic fields with CMB polarization experiments
Authors:
Jostein R. Kristiansen,
Pedro G. Ferreira
Abstract:
We calculate the effect that a primordial homogeneous magnetic field, $\B_0$, will have on the different CMB power spectra due to Faraday rotation. Concentrating on the $TB$, $EB$ and $BB$ correlations, we forecast the ability for future CMB polarization experiments to constrain $\B_0$. Our results depend on how well the foregrounds can be subtracted from the CMB maps, but we find a predicted er…
▽ More
We calculate the effect that a primordial homogeneous magnetic field, $\B_0$, will have on the different CMB power spectra due to Faraday rotation. Concentrating on the $TB$, $EB$ and $BB$ correlations, we forecast the ability for future CMB polarization experiments to constrain $\B_0$. Our results depend on how well the foregrounds can be subtracted from the CMB maps, but we find a predicted error between $σ_{\B_0} = 4 \times 10^{-11}$Gauss (for the QUIET experiment with foregrounds perfectly subtracted) and $3 \times 10^{-10}$Gauss (with the Clover experiment with no foreground subtraction). These constraints are two orders of magnitudes better than the present limits on $\B_0$.
△ Less
Submitted 23 May, 2008; v1 submitted 21 March, 2008;
originally announced March 2008.
-
Cosmological implications of the KATRIN experiment
Authors:
Jostein R. Kristiansen,
Oystein Elgaroy
Abstract:
The upcoming Karlsruhe Tritium Neutrino (KATRIN) experiment will put unprecedented constraints on the absolute mass of the electron neutrino, $\mnue$. In this paper we investigate how this information on $\mnue$ will affect our constraints on cosmological parameters. We consider two scenarios; one where $\mnue=0$ (i.e., no detection by KATRIN), and one where $\mnue=0.3$eV. We find that the const…
▽ More
The upcoming Karlsruhe Tritium Neutrino (KATRIN) experiment will put unprecedented constraints on the absolute mass of the electron neutrino, $\mnue$. In this paper we investigate how this information on $\mnue$ will affect our constraints on cosmological parameters. We consider two scenarios; one where $\mnue=0$ (i.e., no detection by KATRIN), and one where $\mnue=0.3$eV. We find that the constraints on $\mnue$ from KATRIN will affect estimates of some important cosmological parameters significantly. For example, the significance of $n_s<1$ and the inferred value of $Ω_Λ$ depend on the results from the KATRIN experiment.
△ Less
Submitted 2 January, 2008; v1 submitted 26 September, 2007;
originally announced September 2007.
-
Constraining Dark Energy Anisotropic Stress
Authors:
D. F. Mota,
J. R. Kristiansen,
T. Koivisto,
N. E. Groeneboom
Abstract:
We investigate the possibility of using cosmological observations to probe and constrain an imperfect dark energy fluid. We consider a general parameterization of the dark energy component accounting for an equation of state, speed of sound and viscosity. We use present and future data from the cosmic microwave background radiation (CMB), large scale structures and supernovae type Ia. We find th…
▽ More
We investigate the possibility of using cosmological observations to probe and constrain an imperfect dark energy fluid. We consider a general parameterization of the dark energy component accounting for an equation of state, speed of sound and viscosity. We use present and future data from the cosmic microwave background radiation (CMB), large scale structures and supernovae type Ia. We find that both the speed of sound and viscosity parameters are difficult to nail down with the present cosmological data. Also, we argue that it will be hard to improve the constraints significantly with future CMB data sets. The implication is that a perfect fluid description might ultimately turn out to be a phenomenologically sufficient description of all the observational consequences of dark energy. The fundamental lesson is however that even then one cannot exclude, by appealing to observational evidence alone, the possibility of imperfectness in dark energy.
△ Less
Submitted 4 September, 2007; v1 submitted 6 August, 2007;
originally announced August 2007.
-
Using the cluster mass function from weak lensing to constrain neutrino masses
Authors:
Jostein R. Kristiansen,
Oystein Elgaroy,
Hakon Dahle
Abstract:
We discuss the variation of cosmological upper bounds on $M_ν$, the sum of the neutrino masses, with the choice of data sets included in the analysis, pointing out a few oddities not easily seen when all data sets are combined. For example, the effect of applying different priors varies significantly depending on whether we use the power spectrum from the 2dFGRS or SDSS galaxy survey. A conserva…
▽ More
We discuss the variation of cosmological upper bounds on $M_ν$, the sum of the neutrino masses, with the choice of data sets included in the analysis, pointing out a few oddities not easily seen when all data sets are combined. For example, the effect of applying different priors varies significantly depending on whether we use the power spectrum from the 2dFGRS or SDSS galaxy survey. A conservative neutrino mass limit of $M_ν< 1.43$eV (95%C.L.) is obtained by combining the WMAP 3 year data with the cluster mass function measured by weak gravitational lensing. This limit has the virtue of not making any assumptions about the bias of luminous matter with respect to the dark matter, and is in this sense (and this sense only) bias-free.
△ Less
Submitted 10 April, 2007; v1 submitted 24 November, 2006;
originally announced November 2006.
-
Revised WMAP constraints on neutrino masses and other extensions of the minimal $Λ$CDM model
Authors:
Jostein R. Kristiansen,
Hans Kristian Eriksen,
Oystein Elgaroy
Abstract:
Recently, two issues concerning the three-year WMAP likelihood code were pointed out. On large angular scales ($l \lesssim 30$), a sub-optimal likelihood approximation resulted in a small power excess. On small angular scales ($l \gtrsim 300$), over-subtraction of unresolved point sources produced a small power deficit. For a minimal six-parameter cosmological model, these two effects conspired…
▽ More
Recently, two issues concerning the three-year WMAP likelihood code were pointed out. On large angular scales ($l \lesssim 30$), a sub-optimal likelihood approximation resulted in a small power excess. On small angular scales ($l \gtrsim 300$), over-subtraction of unresolved point sources produced a small power deficit. For a minimal six-parameter cosmological model, these two effects conspired to decrease the value of $n_s$ by $\sim 0.7 σ$. In this paper, we study the change in preferred parameter ranges for more extensive cosmological models, including running of $n_s$, massive neutrinos, curvature, and the equation of state for dark energy. We also include large-scale structure and supernova data in our analysis. We find that the parameter ranges for $α_s$, $Ω_k$ and $w$ are not much altered by the modified analysis. For massive neutrinos the upper limit on the sum of the neutrino masses decreases from $M_ν< 1.90$eV to $M_ν< 1.57$eV when using the modified WMAP code and WMAP data only. We also find that the shift of $n_s$ to higher values is quite robust to these extensions of the minimal cosmological model.
△ Less
Submitted 19 December, 2006; v1 submitted 1 August, 2006;
originally announced August 2006.
