Filters








28,844 Hits in 7.4 sec

Modeling Score Distributions and Continuous Covariates: A Bayesian Approach [article]

Mel McCurrie, Hamish Nicholson, Walter J. Scheirer, Samuel Anthony
2020 arXiv   pre-print
We use mixture models to capture arbitrary distributions and local basis functions to capture non-linear, multivariate trends.  ...  We develop a generative model of the match and non-match score distributions over continuous covariates and perform inference with modern Bayesian methods.  ...  Best-Rowden and Jain [3] modeled the effects of elapsed time, race, gender, and other covariates on the match score.  ... 
arXiv:2009.09583v1 fatcat:s4g3ni2ycfcftbdd237xk5bsca

Variational bayes for modeling score distributions

Keshi Dai, Evangelos Kanoulas, Virgil Pavlu, Javed A. Aslam
2010 Information retrieval (Boston)  
In this work, we propose modeling the relevant documents' scores by a mixture of Gaussians and the non-relevant scores by a Gamma distribution.  ...  We test our model on traditional retrieval functions and actual search engines submitted to TREC. We demonstrate the utility of our model in inferring precision-recall curves.  ...  Acknowledgments We would like to thank Avi Arampatzis, Jaap Kamps and Stephen Robertson for many useful discussions.  ... 
doi:10.1007/s10791-010-9156-2 fatcat:3sg2i3ggbbbkrpyjiigk2joagi

Covariate Distribution Balance via Propensity Scores [article]

Pedro H. C. Sant'Anna, Xiaojun Song, Qi Xu
2020 arXiv   pre-print
Heuristically, our proposed procedure attempts to estimate a propensity score model by making the underlying covariate distribution of different treatment groups as close to each other as possible.  ...  This paper proposes new estimators for the propensity score that aim to maximize the covariate distribution balance among different treatment groups.  ...  propensity score model.  ... 
arXiv:1810.01370v4 fatcat:gwwvhgevl5fbrkbeveqemx4jqe

An Iterative Method for Calculating Robust Rating Scores

Mohammad Allahbakhsh, Aleksandar Ignjatovic
2015 IEEE Transactions on Parallel and Distributed Systems  
In this paper we introduce an iterative voting algorithm and then use it to obtain a rating method which is very robust against collusion attacks as well as random and biased raters.  ...  Unlike the previous iterative methods, our method is not based on comparing submitted evaluations to an approximation of the final rating scores, and it entirely decouples credibility assessment of the  ...  INTRODUCTION Human computation is a new model of distributed computing [4, 42] in which the computational power of machines is augmented by the cognitive power of human beings.  ... 
doi:10.1109/tpds.2013.215 fatcat:txuyac3zo5a3dprpio4ydzgc24

Evaluating Models for Model-Based Debugging

Wolfgang Mayer, Markus Stumptner
2008 2008 23rd IEEE/ACM International Conference on Automated Software Engineering  
An empirical comparison is presented that investigates the relative accuracy of different models on a set of test programs and fault assumptions, showing that our abstract interpretation based model provides  ...  We provide an overview of current techniques in model-based debugging and assess strengths and weaknesses of the individual approaches.  ...  Guided exploration of fault assumptions results in robust explanations that remain close to the original correct model.  ... 
doi:10.1109/ase.2008.23 dblp:conf/kbse/MayerS08 fatcat:uckvwytmx5axhfnkqd6czeqbrm

Accelerated failure time models provide a useful statistical framework for aging research

William R. Swindell
2009 Experimental Gerontology  
AFT model deceleration factors also provided a more intuitive measure of treatment effect than the hazard ratio, and were robust to departures from modeling assumptions.  ...  Most genetic manipulations were found to have a multiplicative effect on survivorship that is independent of age and well-characterized by the AFT model "deceleration factor".  ...  Rabinovitch, Qitao Ran, Arlan Richardson and Junji Yodoi.  ... 
doi:10.1016/j.exger.2008.10.005 pmid:19007875 pmcid:PMC2718836 fatcat:jrspbc67zranrlcew32bgk2md4

Causal Modeling for Fairness in Dynamical Systems [article]

Elliot Creager, David Madras, Toniann Pitassi, Richard Zemel
2020 arXiv   pre-print
We show that this formulation affords several new directions of inquiry to the modeler, where causal assumptions can be expressed and manipulated.  ...  to produce both immediate and long-term effects for individuals and demographic groups.  ...  Acknowledgements We thank Kuan-Chieh Wang, Eleni Triantafillou, Will Grathwohl, James Lucas, Robert Adragna, and John Miller for providing feedback, and the reviewers for their helpful suggestions.  ... 
arXiv:1909.09141v2 fatcat:kt33xsq3tffvvoigxgvj4ngcme

Adversarial Deep Learning Models with Multiple Adversaries

Aneesh Chivukula, Wei Liu
2018 IEEE Transactions on Knowledge and Data Engineering  
We then retrain the learner over all the adversarial manipulations generated by multiple players to propose a secure CNN which is robust to subsequent adversarial data manipulations.  ...  The algorithm's objective is to produce small changes to the data distribution defined over positive and negative class labels so that the resulting data distribution is misclassified by the CNN.  ...  INTRODUCTION To learn mathematical patterns, machine learning methods make assumptions on the data distributions for training and testing the learning algorithm.  ... 
doi:10.1109/tkde.2018.2851247 fatcat:vc3xdxrswrhefoc3oam7uaot6e

Comparing Models of Change to Estimate the Mediated Effect in the Pretest–Posttest Control Group Design

Matthew J. Valente, David P. MacKinnon
2017 Structural Equation Modeling  
score, and cross-sectional model.  ...  When these assumptions do not hold, Type I error and statistical power results suggest that only the ANCOVA model has good performance. The four models are applied to an empirical example.  ...  Acknowledgments The authors would like to acknowledge Leona Aiken, Craig Enders, and Stephen West for their invaluable feedback on the ideas presented in this manuscript.  ... 
doi:10.1080/10705511.2016.1274657 pmid:28845097 pmcid:PMC5568008 fatcat:sycv7fzw3zevdafb5ssxx7lgry

Modeling grid fields instead of modeling grid cells [article]

Sophie Rosay, Simon N. Weber, Marcello Mulas
2018 bioRxiv   pre-print
To do so, we use an effective model and consider grid fields as point objects (particles) moving around in space under the influence of forces.  ...  Finally, we draw the relationship between our 'macroscopic' model for grid fields and existing 'microscopic' models of grid cell activity and discuss how a description at the level of grid fields allows  ...  Acknowledgements We are grateful to Alessandro Treves, Rémi Monasson, Giuseppe D'Adamo, Thomas Gueudré and Henning Sprekeler for their remarks on the model and its relationship with Physics.  ... 
doi:10.1101/481747 fatcat:orqgybebdbcijfjr2au5javu5i

Linkage mapping in experimental crosses: the robustness of single-gene models

F A Wright, A Kong
1997 Genetics  
By examining the expected LOD across the genome, it is shown that single-gene models are quite robust, even for polygenic traits.  ...  The results are derived for backcross populations, with a discussion of extensions to intercross populations and relative-pair mapping in humans.  ...  The authors thank the editor and referees for valuable suggestions.  ... 
pmid:9136030 pmcid:PMC1207957 fatcat:rxyv5zjo6zdjfi54cjonxwj7le

Robustness Metric for Quantifying Causal Model Confidence and Parameter Uncertainty [article]

Garrett Waycaster, Christian Bes, Volodymyr Bilotkach, Christian Gogu, Raphael Haftka, Nam-Ho Kim
2016 arXiv   pre-print
The use of this metric is demonstrated on both numerically simulated data and a case study from existing causal model literature.  ...  The metric is developed for time series causal models, but is also applicable to non-time series data.  ...  ACKNOWLEDGEMENTS The authors gratefully acknowledge the support of the National Science Foundation grants CMMI-0927790 and 1131103 which funded this work.  ... 
arXiv:1602.02198v1 fatcat:vquovsmzunbkhblc2546lxkk4m

Perfect density models cannot guarantee anomaly detection [article]

Charline Le Lan, Laurent Dinh
2021 arXiv   pre-print
We conclude that the use of these likelihoods for out-of-distribution detection relies on strong and implicit hypotheses, and highlight the necessity of explicitly formulating these assumptions for reliable  ...  Thanks to the tractability of their likelihood, some deep generative models show promise for seemingly straightforward but important applications like anomaly detection, uncertainty estimation, and active  ...  We would also like to thank the Python community (Van Rossum and Drake Jr, 1995; Oliphant, 2007) for developing tools that enabled this work, including NumPy (Oliphant, 2006; Walt et al., 2011; Harris  ... 
arXiv:2012.03808v2 fatcat:kchdoxof6bhbxm6k3cnbopjuum

Improved Query Topic Models via Pseudo-Relevant Pólya Document Models [article]

Ronan Cummins
2016 arXiv   pre-print
likely to belong to the query topic model.  ...  We assume that documents are generated via a mixture of multivariate Polya distributions, and we show that by identifying the topical terms in each document, we can appropriately select terms that are  ...  model and p(q|θ d ) is the query likelihood score (document score) 1 .  ... 
arXiv:1602.01665v1 fatcat:susa6czxcfak5mozdbi7oqjbh4

Modeling Representation of Minorities Under Multiwinner Voting Rules [article]

Piotr Faliszewski and Jean-Francois Laslier and Robert Schaefer and Piotr Skowron and Arkadii Slinko and Nimrod Talmon
2016 arXiv   pre-print
First and foremost, we aim to experimentally and quantitatively show that the choice of a multiwinner voting rule can play a crucial role on the way minorities are represented.  ...  We also test the possibility for some of these rules to achieve proportional representation.  ...  The assumption that the set of candidates is identical to the set of voters is often met in the Political Economy literature since [14, 2] and labelled the "citizen-candidate" model. 4 "Why being centrist  ... 
arXiv:1604.02364v1 fatcat:whypo5uexvd67bae3wq6mslmwe
« Previous Showing results 1 — 15 out of 28,844 results