Filters








22,208 Hits in 8.7 sec

Reliable Post hoc Explanations: Modeling Uncertainty in Explainability [article]

Dylan Slack, Sophie Hilgard, Sameer Singh, Himabindu Lakkaraju
2021 arXiv   pre-print
As black box explanations are increasingly being employed to establish model credibility in high-stakes settings, it is important to ensure that these explanations are accurate and reliable.  ...  We carry out a detailed theoretical analysis that leverages the aforementioned uncertainty to estimate how many perturbations to sample, and how to sample for faster convergence.  ...  The views expressed are those of the authors and do not reflect the official policy or position of the funding agencies.  ... 
arXiv:2008.05030v4 fatcat:ssbibsmworf2bpm7rye4wpaevq

How Much Can I Trust You? – Quantifying Uncertainties in Explaining Neural Networks [article]

Kirill Bykov, Marina M.-C. Höhne, Klaus-Robert Müller, Shinichi Nakajima, Marius Kloft
2020 arXiv   pre-print
model uncertainties into a quantification of explanation uncertainties.  ...  modeling of uncertainties.  ...  Figure 10 : 10 Another exemplary visualization of the clever hans effect for an image depicting a horse. arXiv:2006.09000v1 [cs.LG] 16 Jun 2020 Preprint -How Much Can I Trust You?  ... 
arXiv:2006.09000v1 fatcat:s7262yb2pzfofe5qgcq742uogy

Can You Trust Your Trust Measure? [article]

Meia Chita-Tegmark, Theresa Law, Nicholas Rabb, Matthias Scheutz
2021 arXiv   pre-print
and the exploration of social and relational dimensions of trust in robots (e.g., benevolence).  ...  In this paper we look at how different trust questionnaires fare given these challenges that pull in different directions (being general vs. being exploratory) by studying whether people think the items  ...  I promise I won't immobilize you." [here, the picture of the robot appears again and stays for the remainder of the vignette]. Then you start playing.  ... 
arXiv:2104.11365v1 fatcat:jlgrd5wp2vh3fbwmt4uubmwo5i

Tell Me Something That Will Help Me Trust You: A Survey of Trust Calibration in Human-Agent Interaction [article]

George J. Cancro, Shimei Pan, James Foulds
2022 arXiv   pre-print
In this paper we survey literature in the area of trust between a single human supervisor and a single agent subordinate to determine the nature and extent of this additional information and to characterize  ...  agent to decide whether to trust or reject the prediction or recommendation?  ...  from a black box agent [Guidotti et al., 2019] .  ... 
arXiv:2205.02987v1 fatcat:6imqgkctprayrnpift2ybwc7u4

"I can assure you [...] that it's going to be all right" -- A definition, case for, and survey of algorithmic assurances in human-autonomy trust relationships [article]

Brett W Israelsen
2017 arXiv   pre-print
In essence they want to be able to trust the systems that are being designed. In this survey we present assurances that are the method by which users can understand how to trust this technology.  ...  Trust between humans and autonomy is reviewed, and the implications for the design of assurances are highlighted.  ...  ects, black-box models are of li le use.  ... 
arXiv:1708.00495v2 fatcat:bx42oqcijfhchna3xhl7nktnqa

"Dave...I can assure you ...that it's going to be all right ..." A Definition, Case for, and Survey of Algorithmic Assurances in Human-Autonomy Trust Relationships

Brett W. Israelsen, Nisar R. Ahmed
2019 ACM Computing Surveys  
Algorithmic assurances are first formally defined and classified from the perspective of formally modeled human-artificially intelligent agent trust relationships.  ...  However, these approaches are typically ad hoc, and have not been formally related to each other or to formal trust models.  ...  How should explanations be provided?  ... 
doi:10.1145/3267338 fatcat:7wzpdawvwbdmvavttqfsu2fly4

"Now You See Me, Now You Don't": How Digital Consumers Manage Their Online Visibility in Game-Like Conditions

Rikke Duus, Mike Cooray, Simon Lilley
2022 Frontiers in Psychology  
In this study, we draw upon Erving Goffman's metaphor of expression games and three forms of concealment or cover moves to explore how consumers, who have been well socialized as digital natives, engage  ...  Our work contributes with new insight into how these consumers actively engage in the orchestration of their online visibility by surfacing the nuanced and multifaceted decision-making and thought processes  ...  You very consciously uncheck boxes now. Whereas before, you'd be like, "Yeah, tick it and leave it." But now you actually read what they're asking.  ... 
doi:10.3389/fpsyg.2022.795264 fatcat:dvaglvjcenej5gz5h5hnjqyf2i

Shall I Trust You? From Child–Robot Interaction to Trusting Relationships

Cinzia Di Dio, Federico Manzi, Giulia Peretti, Angelo Cangelosi, Paul L. Harris, Davide Massaro, Antonella Marchetti
2020 Frontiers in Psychology  
The relationship between trust and the representation of the quality of attachment relationships, Theory of Mind, and executive function skills was also investigated.  ...  Studying trust in the context of human-robot interaction is of great importance given the increasing relevance and presence of robotic agents in the social sphere, including educational and clinical.  ...  Mario Uboldi, Principal of the "I.C. Giovanni Pascoli", and Mrs. Maura Panzone for their important contribution to the organization of the project and support to the study.  ... 
doi:10.3389/fpsyg.2020.00469 pmid:32317998 pmcid:PMC7147504 fatcat:guirpockcfhb3cvcm7cythshim

"Wish you were here" trust in public administration in Latin America
"Wish you were here" confianza en la administración pública en Latinoamérica

Cecilia Güemes
2019 Revista de Administração Pública  
Question: How much do you trust the following professionals? Scale from 1 to 5 where 1 is not confident at all and 5 is full trust.  ...  Question: How much do you trust the following professionals? Scale from 1 to 5 where 1 is not confi dent at all and 5 is full trust. Source: CAF (2014).  ... 
doi:10.1590/0034-761220180230x fatcat:qggumkgeebeejlcvquqwtvlcnm

Slave to the Algorithm? Why a Right to Explanationn is Probably Not the Remedy You are Looking for

Lilian Edwards, Michael Veale
2017 Social Science Research Network  
Transparency in the form of a "right to an explanation" has emerged as a compellingly attractive remedy since it intuitively promises to open the algorithmic "black box" to promote challenge, redress,  ...  However, "subject-centric" explanations (SCEs) focussing on particular regions of a model around a query show promise for interactive exploration, as do explanation systems based on learning a model from  ...  "Why Should I Trust You?"  ... 
doi:10.2139/ssrn.2972855 fatcat:nqqo4xmminc4bkp76mwnrbhgeu

What You Sow Is What You Reap? (Dis-)Incentives for Adaptation Intentions in Farming

Franca Buelow, Nicholas Cradock-Henry
2018 Sustainability  
We peer into the black box of adaptation intention to look more closely at the relationship between financial incentives and adaptation action, in order to provide new insights to policy-makers and practitioners  ...  Climate change effects are quite heterogeneous and involve a high degree of uncertainty, which further complicates planning.  ...  Trust in technology is the first step of application, and the use of such technologies precedes societal support and acceptance.  ... 
doi:10.3390/su10041133 fatcat:wt5kifld55adpgoehyz36zu7su

'You can dance your prototype if you like': independent filmmakers adapting the hackathon

Joakim Karlsen, Anders Sundnes Løvlie
2017 Digital Creativity  
And of course a boss that trusts you a hundred per cent, because time is ticking... ".  ...  They conveyed that they had little appetite for opening the "black box" of technology production and thought that they could outsource this work if necessary.  ...  How facilitating for audience participation, have challenged earlier practices of making independent documentary  ... 
doi:10.1080/14626268.2017.1351992 fatcat:3k2jwoz2tjfypntlnskhjslkri

Six (Or So) Things You Can Do with a Bad Model

James S. Hodges
1991 Operations Research  
This paper takes a step toward providing such a theory b, focusing on the logic that should govern the use of inadequatel, validated models and the costs and benefits of using them.  ...  For example, in the area of combat anal)sis. the central models are arguably almost entirely unvalidated and most %kill never be susceptible to adequate validation.  ...  It is tool ('here is how much mones you t ill save if you hard to imagine an instance in which this will not put the make the tradeoff this other waY').  ... 
doi:10.1287/opre.39.3.355 fatcat:mts7ze4bxrbbvi2j4ynp2insvu

"One more time: how do you win orders?": a critical reappraisal of the Hill manufacturing strategy framework

Martin Spring, Ruth Boaden
1997 International Journal of Operations & Production Management  
(4) Do developments in the strategy literature -notably the core competence view -make the basis of much of the manufacturing strategy literature, including Hill's[1] approach, obsolete?  ...  (2) Is the order winner/order qualifier really different from the competitive criteria or competitive priority concept often adopted in the manufacturing strategy literature, and how do they compare to  ...  in that a high level of trust can greatly mitigate potentially problematic uncertainty.  ... 
doi:10.1108/01443579710175547 fatcat:4tv4nb33qzef5omkcp2jkownv4

Facing Fault Management as It Is, Aiming for What You Would Like It to Be [chapter]

Roy Sterritt
2002 Lecture Notes in Computer Science  
Fault management is a complex task, subject to uncertainty in the symptoms presented.  ...  In this paper two key fault management approaches are considered: (i) rule discovery to attempt to present fewer symptoms with greater diagnostic assistance for the more traditional rule based system approach  ...  The paper has benefited from discussions with other members of the Jigsaw team and with collaborators at Nortel.  ... 
doi:10.1007/3-540-46019-5_3 fatcat:3nizbbjohvh3ranie5ydswv4hy
« Previous Showing results 1 — 15 out of 22,208 results