Filters








784 Hits in 6.2 sec

Mitigating belief projection in explainable artificial intelligence via Bayesian Teaching [article]

Scott Cheng-Hsin Yang, Wai Keen Vong, Ravi B. Sojitra, Tomas Folke, Patrick Shafto
2021 arXiv   pre-print
We assess Bayesian Teaching in a binary image classification task across a variety of contexts.  ...  We propose explicitly modeling the human explainee via Bayesian Teaching, which evaluates explanations by how much they shift explainees' inferences toward a desired goal.  ...  Supplementary information: Mitigating belief projection in explainable artificial intelligence via Bayesian Teaching Supplementary 80 participants (25 male, 54 female, 1 other) were recruited from Amazon  ... 
arXiv:2102.03919v2 fatcat:oviy2bdskfdfbgxvhu6ne4a64a

Mitigating belief projection in explainable artificial intelligence via Bayesian teaching

Scott Cheng-Hsin Yang, Wai Keen Vong, Ravi B. Sojitra, Tomas Folke, Patrick Shafto
2021 Scientific Reports  
We assess Bayesian teaching in a binary image classification task across a variety of contexts.  ...  We propose explicitly modelling the human explainee via Bayesian teaching, which evaluates explanations by how much they shift explainees' inferences toward a desired goal.  ...  Our results imply that such belief-projection can be mitigated by Bayesian teaching.  ... 
doi:10.1038/s41598-021-89267-4 pmid:33972625 fatcat:t7dcl37vpzckvpdjq4esrccg7i

Explainable AI for Natural Adversarial Images [article]

Tomas Folke, ZhaoBin Li, Ravi B. Sojitra, Scott Cheng-Hsin Yang, Patrick Shafto
2021 arXiv   pre-print
In previous work we have found that humans tend to assume that the AI's decision process mirrors their own.  ...  Human oversight might mitigate this weakness, but depends on humans understanding the AI well enough to predict when it is likely to make a mistake.  ...  Mitigating belief projection in explainable artificial intelligence via bayesian teaching.  ... 
arXiv:2106.09106v1 fatcat:huvnzephw5bv5fo5mq2rja76ri

Assessing Risks of Biases in Cognitive Decision Support Systems [article]

Kenneth Lai, Helder C. R. Oliveira, Ming Hou, Svetlana N. Yanushkevich, Vlad Shmerko
2020 arXiv   pre-print
Recognizing, assessing, countering, and mitigating the biases of different nature from heterogeneous sources is a critical problem in designing a cognitive Decision Support System (DSS).  ...  We provide performance projections of the DSS operational landscape in terms of biases. A probabilistic reasoning technique is used for assessment of the risk of such biases.  ...  ACKNOWLEDGMENTS This Project was partially supported by Natural Sciences and Engineering Research Council of Canada (NSERC) through grant "Biometric-Enabled Identity Management and Risk Assessment for  ... 
arXiv:2007.14361v1 fatcat:g36n75gr7jbplkrpzmnshug2h4

Risk, Trust, and Bias: Causal Regulators of Biometric-Enabled Decision Support

Kenneth Lai, Helder C. R. Oliveira, Ming Hou, Svetlana N. Yanushkevich, Vlad P. Shmerko
2020 IEEE Access  
Practical details of the R-T-B assessment in the DSS are demonstrated using the experiments of assessing the trust in synthetic biometric and the risk of bias in face biometrics.  ...  The proposed novel taxonomy links the R-T-B assessment to the causal inference mechanism for reasoning in decision making.  ...  Responses to the algorithmic bias, in particular, include identifying and intervening problematic biases. A related bias is known as the artificial intelligence bias [77] .  ... 
doi:10.1109/access.2020.3015855 pmid:34812347 pmcid:PMC8545314 fatcat:eus6hlbgyzfhnk3fcu4zfxxksi

A Bayesian social platform for inclusive and evidence-based decision making [article]

Susannah Kate Devitt, Tamara Rose Pearce, Alok Kumar Chowdhury, Kerrie Mengersen
2021 arXiv   pre-print
A Bayesian social network can make virtuous ideas powerful.  ...  Social media success, we argue, is in the hypothesis. Documents are valuable to the degree that they are evidence in service of, or to challenge an idea for a purpose.  ...  intelligence.  ... 
arXiv:2102.06893v1 fatcat:eygevbiobvbtrdz3bnzlxlabom

Risk, Trust, and Bias: Causal Regulators of Biometric-Enabled Decision Support [article]

Kenneth Lai, Helder C. R. Oliveira, Ming Hou, Svetlana N. Yanushkevich, Vlad P. Shmerko
2020 arXiv   pre-print
Practical details of the R-T-B assessment in the DSS are demonstrated using the experiments of assessing the trust in synthetic biometric and the risk of bias in face biometrics.  ...  The proposed novel taxonomy links the R-T-B assessment to the causal inference mechanism for reasoning in decision making.  ...  ACKNOWLEDGMENTS This Project was partially supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) through grant "Biometric-enabled Identity management and Risk Assessment  ... 
arXiv:2008.02359v1 fatcat:tnzoptk26fazhhjfmvrm2zprbe

Emerging Biometrics: Deep Inference and Other Computational Intelligence [article]

Svetlana Yanushkevich, Shawn Eastwood, Kenneth Lai, Vlad Shmerko
2020 arXiv   pre-print
Therefore, we focus on intelligent inference engines widely deployed in biometrics.  ...  We highlight the technology gaps that must be addressed in future generations of biometric systems.  ...  Acknowledgment This project was partially supported by Natural Sciences and Engineering Research Council of Canada (NSERC) through the grant "Biometric intelligent interfaces".  ... 
arXiv:2006.11971v1 fatcat:k6aunuoxc5apbb347vy2lj7bsq

Theory of Mind and Preference Learning at the Interface of Cognitive Science, Neuroscience, and AI: A Review

Christelle Langley, Bogdan Ionut Cirstea, Fabio Cuzzolin, Barbara J. Sahakian
2022 Frontiers in Artificial Intelligence  
The same capability of inferring human mental states is a prerequisite for artificial intelligence (AI) to be integrated into society, for example in healthcare and the motoring industry.  ...  In the literature, there has been an increasing understanding of ToM, specifically with increasing cognitive science studies in children and in individuals with Autism Spectrum Disorder.  ...  ACKNOWLEDGMENTS BC acknowledges the Alignment Newsletter, which helped significantly with surveying preference ToM algorithms in AI.  ... 
doi:10.3389/frai.2022.778852 pmid:35493614 pmcid:PMC9038841 fatcat:bqemggi3wverbmizmae7posxqi

Intelligence Primer [article]

Karl Fezer, Andrew Sloss
2022 arXiv   pre-print
Intelligence is a fundamental part of all living things, as well as the foundation for Artificial Intelligence.  ...  Artificial Intelligence, in the form of Machine Learning, has already had a significant impact on our lives. As an exploration, we journey into different parts of intelligence that appear essential.  ...  Bayesian probabilism is underpinned by the Bayes Theorem, which in practical terms is a mathematical equation for how justified a specific belief is about the world.  ... 
arXiv:2008.07324v3 fatcat:not5i5utwzfzboi42yctekn7sy

Adaptive Robot Language Tutoring Based on Bayesian Knowledge Tracing and Predictive Decision-Making

Thorsten Schodde, Kirsten Bergmann, Stefan Kopp
2017 Proceedings of the 2017 ACM/IEEE International Conference on Human-Robot Interaction - HRI '17  
Being implemented in a robot language tutor, the model enables the robot tutor to trace the learner's knowledge and to decide which skill to teach next and how to address it in a game-like tutoring interaction  ...  In this paper, we present an approach to adaptive language tutoring in child-robot interaction.  ...  This is realized via an extended model that combines Bayesian Knowledge Tracing (of the learned) with tutoring actions (of the tutor) in one causal probabilistic model.  ... 
doi:10.1145/2909824.3020222 dblp:conf/hri/SchoddeBK17 fatcat:vism2vgfojae5krlodxj2n42ai

AGI Safety Literature Review

Tom Everitt, Gary Lea, Marcus Hutter
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
The development of Artificial General Intelligence (AGI) promises to be a major event. Along with its many potential benefits, it also raises serious safety concerns.  ...  the study of artificial intelligence.  ...  Introduction An Artificial General Intelligence (AGI) is an AI system that equals or exceeds human intelligence in a wide variety of cognitive tasks.  ... 
doi:10.24963/ijcai.2018/768 dblp:conf/ijcai/EverittLH18 fatcat:3z5wqjw5mvhstik5bmdhhu2gnq

Back to the Basics: Bayesian extensions of IRT outperform neural networks for proficiency estimation [article]

Kevin H. Wilson, Yan Karklin, Bojian Han, Chaitanya Ekanadham
2016 arXiv   pre-print
When data sets included non-trivial autocorrelations in student response patterns, a temporal extension of IRT improved performance over standard IRT while the RNN-based method did not.  ...  models provide a simpler, better-performing alternative to existing RNN-based models of student interaction data while also affording more interpretability and guarantees due to their formulation as Bayesian  ...  Each µj is in turn distributed according to the hyperprior µj ∼ N (0, τ 2 ). This reflects the belief that the difficulty of items in the same group should be similar.  ... 
arXiv:1604.02336v2 fatcat:v77xnwzf5fc55fx6n3zn3vl5dm

Foundations of Explainable Knowledge-Enabled Systems [article]

Shruthi Chari, Daniel M. Gruen, Oshani Seneviratne, Deborah L. McGuinness
2020 arXiv   pre-print
Explainability has been an important goal since the early days of Artificial Intelligence. Several approaches for producing explanations have been developed.  ...  We present a historical overview of explainable artificial intelligence systems, with a focus on knowledge-enabled systems, spanning the expert systems, cognitive assistants, semantic applications, and  ...  explainable-artificial-intelligence Bau et. al  ... 
arXiv:2003.07520v1 fatcat:pz54e4ag35hf3osr7dfbmhb4ze

AGI Safety Literature Review [article]

Tom Everitt, Gary Lea, Marcus Hutter
2018 arXiv   pre-print
The development of Artificial General Intelligence (AGI) promises to be a major event. Along with its many potential benefits, it also raises serious safety concerns (Bostrom, 2014).  ...  top 100 most cited authors in artificial intelligence.  ...  the study of artificial intelligence.  ... 
arXiv:1805.01109v2 fatcat:v7vno74ngrcpljfj2exy5vf7rq
« Previous Showing results 1 — 15 out of 784 results