178,448 Hits in 3.6 sec

Hazards and fallacies of statistical method in psychological measurement

John Gray Peatman
1937 The Psychological Record  
Statistical inference, which might be defined as reasoning from samples, is a form of probable 378 JOHN GRAY PEATMAN inference, in contrast to necessary inference.  ...  In the second place, the answer to our question is no because of the very nature of statistical inference.  ... 
doi:10.1007/bf03393206 fatcat:jc75fbusdbelxl7m7ptczbonnm

Exploiting Fairness to Enhance Sensitive Attributes Reconstruction [article]

Julien Ferry
2022 arXiv   pre-print
To assess the applicability of our approach, we have conducted a thorough experimental evaluation on two state-of-the-art fair learning methods, using four different fairness metrics with a wide range  ...  The proposed method is agnostic to the type of target model, the fairness-aware learning method as well as the auxiliary knowledge of the adversary.  ...  Three main approaches have emerged [20] , namely statistical fairness metrics, individual fairness and causal fairness.  ... 
arXiv:2209.01215v1 fatcat:32cyuc2cszfkrn4x5o35tivfue

Why People are Not Like Marbles in an Urn: An Effect of Context on Statistical Reasoning

1996 Applied Cognitive Psychology  
Relatively little work, however, has examined context effects on statistical  ...  A prototheory of fairness, for example, may be a precursor for a theory of stratification (e.g., chose 25 boys and 25 girls).  ...  The fairness prototheory does not permit an evaluation of the effects of self-selection on a statistical inference (e.g., maybe only the students who want to go to the fun booth would bother to answer  ... 
doi:10.1002/(sici)1099-0720(199611)10:7<99::aid-acp441>;2-4 fatcat:qouu43puonfslnglvlrpt5kxfm

Promises and Challenges of Causality for Ethical Machine Learning [article]

Aida Rahmattalabi, Alice Xiang
2022 arXiv   pre-print
We highlight key aspects of causal inference that are often ignored in the causal fairness literature.  ...  Subsequently, we illustrate how causality can address the limitations of existing fairness metrics, including those that depend upon statistical correlations.  ...  For fairness violation, we consider both statistical criteria and their causal variants. Table 1 summarizes the statistical fairness violation results for β = 1.0 and γ = 0.2.  ... 
arXiv:2201.10683v1 fatcat:lbsirx3pczewjeel22xpwvxtry

An Ontology for Fairness Metrics

Jade S. Franklin, Karan Bhanot, Mohamed Ghalwash, Kristin P. Bennett, Jamie McCusker, Deborah L. McGuinness
2022 Proceedings of the 2022 AAAI/ACM Conference on AI, Ethics, and Society  
However, determining which metrics to use, as well as interpreting their results, is difficult for a non-expert due to a lack of clear guidance and issues of ambiguity or alternate naming schemes between  ...  To address this knowledge gap, we present the Fairness Metrics Ontology (FMO), a comprehensive and extensible knowledge resource that defines each fairness metric, describes their use cases, and details  ...  , and the Statistical Test-Based Fairness Notion, which is satisfied when a statistical test for fairness resolves with a specific p-value.  ... 
doi:10.1145/3514094.3534137 fatcat:yg35za36tzgexinahnrfhmfx2u

A Declarative Approach to Fairness in Relational Domains

Golnoosh Farnadi, Behrouz Babaki, Lise Getoor
2019 IEEE Data Engineering Bulletin  
However, existing fairness approaches are based solely on attributes of individuals.  ...  Furthermore, we extend an existing statistical relational learning framework, probabilistic soft logic (PSL), to incorporate our definition of relational fairness.  ...  For example, fairness through awareness/Lipschitz property [11] , individual fairness [27] , statistical parity/group fairness [17] , counterfactual fairness [19] , demographic parity/disparate impact  ... 
dblp:journals/debu/FarnadiBG19 fatcat:ifbtwahfqjd3tdjedzzhxryloe

The Frontiers of Fairness in Machine Learning [article]

Alexandra Chouldechova, Aaron Roth
2018 arXiv   pre-print
Along the way, it surveys recent theoretical work in the field and points towards promising directions for research.  ...  The last few years have seen an explosion of academic and popular interest in algorithmic fairness.  ...  Individual Definitions of Fairness Individual notions of fairness, on the other hand, ask for constraints that bind on specific pairs of individuals, rather than on a quantity that is averaged over groups  ... 
arXiv:1810.08810v1 fatcat:nf7ddavgsfavxbe5nqgkkxftiu

Fair inference on error-prone outcomes [article]

Laura Boeschoten, Erik-Jan van Kesteren, Ayoub Bagheri, Daniel L. Oberski
2020 arXiv   pre-print
Fair inference in supervised learning is an important and active area of research, yielding a range of useful methods to assess and account for fairness criteria when predicting ground truth targets.  ...  , and, on the other, measurement models found in the statistical literature.  ...  Note that this definition looks superficially similar to the definition of statistical parity (group fairness), but is distinct because it refers to an individual.  ... 
arXiv:2003.07621v1 fatcat:bs5ehzodvjfgfmembcxzoupq7u

Reform options for peer review

R. H. Bradshaw, N. E. Bubier
1993 Nature  
Any teacher or student in higher education will know that there is always going to be a variance between the judgements of individuals where there is a high degree of inference.  ...  Application of a statistical method to analyse a particular dataset is either right or wrong and the 20 referees who thought that the statistical methods were acceptable, fair or good were probably not  ... 
doi:10.1038/364183a0 pmid:8321305 fatcat:fmzwhuk35jeo7iwtydhmadmsbq

Principal Fairness for Human and Algorithmic Decision-Making [article]

Kosuke Imai, Zhichao Jiang
2022 arXiv   pre-print
Unlike the existing statistical definitions of fairness, principal fairness explicitly accounts for the fact that individuals can be impacted by the decision.  ...  Using the concept of principal stratification from the causal inference literature, we introduce a new notion of fairness, called principal fairness, for human and algorithmic decision-making.  ...  for these individuals.  ... 
arXiv:2005.10400v5 fatcat:vlfsvgr5vzh6rmc6zrirjcgtza

Universal adaptability: Target-independent inference that competes with propensity scoring

Michael P. Kim, Christoph Kern, Shafi Goldwasser, Frauke Kreuter, Omer Reingold
2022 Proceedings of the National Academy of Sciences of the United States of America  
The gold-standard approaches for gleaning statistically valid conclusions from data involve random sampling from the population.  ...  Our approach builds on a surprising connection between the problem of inferences in unspecified target populations and the multicalibration problem, studied in the burgeoning field of algorithmic fairness  ...  Inference Task We aim for accurate statistical inferences over a target distribution.  ... 
doi:10.1073/pnas.2108097119 pmid:35046023 pmcid:PMC8794832 fatcat:ivzizi4vk5ccxgpgcwgd3kcowu

Achieving Fair Inference Using Error-Prone Outcomes

Laura Boeschoten, Erik-Jan van Kesteren, Ayoub Bagheri, Daniel L. Oberski
2021 International Journal of Interactive Multimedia and Artificial Intelligence  
found in the statistical literature.  ...  Recently, an increasing amount of research has focused on methods to assess and account for fairness criteria when predicting ground truth targets in supervised learning.  ...  Note that this definition looks superficially similar to the definition of statistical parity (group fairness), but is distinct because it refers to an individual.  ... 
doi:10.9781/ijimai.2021.02.007 doaj:93c30af805304381a69f771fbe001ea9 fatcat:ci27ix6rdnddhktl7mgilydz54

A review of causality-based fairness machine learning

Cong Su, Guoxian Yu, Jun Wang, Zhongmin Yan, Lizhen Cui
2022 Intelligence & Robotics  
The review begins by introducing the common causality-based definitions and measures for fairness.  ...  Early studies focused on defining the correlation/association-based notions, such as statistical parity, equalized odds, etc.  ...  SP is a common statistics-based fairness notion, which denotes similar individuals treated similarly regardless of their sensitive attributes.  ... 
doi:10.20517/ir.2022.17 fatcat:6grnttrnx5hdhdzifhfmhkg6ji

Fairness in Machine Learning with Tractable Models [article]

Michael Varley, Vaishak Belle
2020 arXiv   pre-print
In particular, we show that sum product networks (SPNs) enable an effective technique for determining the statistical relationships between protected attributes and other training variables.  ...  We will also motivate the concept of "fairness through percentile equivalence", a new definition predicated on the notion that individuals at the same percentile of their respective distributions should  ...  Finally, there are also concerns that statistical parity measures fail to account for fair treatment of individuals [12] , a topic we will explore more when considering the shortcomings of fairness through  ... 
arXiv:1905.07026v2 fatcat:54hupbxitfby5f4np47zvaqrtu

Challenges and Opportunities in Statistics and Data Science: Ten Research Areas

Xuming He, Xihong Lin
2020 Harvard Data Science Review  
As a discipline that deals with many aspects of data, statistics is a critical pillar in the rapidly evolving landscape of data science.  ...  Statistics plays a pivotal role in data science by assisting with the use of data and decision making in the face of uncertainty.  ...  Cloud-Based Scalable and Distributed Statistical Inference It is of high importance to develop practical scalable statistical inference for the analysis of real-world massive data.  ... 
doi:10.1162/99608f92.95388fcb fatcat:6wj3zu4dlze6tpnd3govhxscq4
« Previous Showing results 1 — 15 out of 178,448 results