Results 41 to 50 of about 144,405 (196)

Decision Threshold Setting in Binary Classification Problems—A Behavioral Lens

Journal of Operations Management, EarlyView.
ABSTRACT When binary classification models are wrong, managers face misclassification costs. Although false positive outcomes imply unnecessary mitigation efforts, false negative outcomes imply overlooking the class of interest. Humans calibrate these ai models supporting operational systems by adjusting the decision threshold that translates ...
Patrick Moder, Kai Hoberg, Felix Papier
wiley   +1 more source

Preventing False Discovery in Interactive Data Analysis is Hard [PDF]

, 2014
We show that, under a standard hardness assumption, there is no computationally efficient algorithm that given $n$ samples from an unknown distribution can give valid answers to $n^{3+o(1)}$ adaptively chosen statistical queries. A statistical query asks
Hardt, Moritz, Ullman, Jonathan
core   +1 more source

Predicting externalizing symptom trajectories in U.S. National Guard recruits: The role of adverse childhood experiences

Journal of Traumatic Stress, EarlyView.
Abstract Adverse childhood experiences (ACEs) are strongly associated with increased risk of externalizing problems. Despite their prevalence in military populations, limited research links ACEs to longitudinal externalizing problem trajectories during military service transition.
Ali F. Sloan, Tristan Bron, Craig A. Marquardt, Seth G. Disner, Siamak Noorbaloochi, Melissa A. Polusny, Jonathan D. Schaefer   +6 more
wiley   +1 more source

Prior osteosynthesis—unlike osteotomy—raises revision risk after total knee arthroplasty, predominantly via periprosthetic infection

Knee Surgery, Sports Traumatology, Arthroscopy, EarlyView.
Abstract Purpose Previous knee surgery is discussed as a risk factor for total knee arthroplasty (TKA), yet most reports do not distinguish between osteotomy and osteosynthesis. The German Operationen‐ und Prozedurenschlüssel (OPS)‐based classification was used to compare the impact of these procedures on implant survival, mortality and complications ...
Nele Wagener, Yinan Wu, Alexander Grimberg, Christian Hipfl, Sebastian Hardt   +4 more
wiley   +1 more source

Crisis of Law of Value and the Real Subsumption of Labor under Capital. Questions for Michael Hardt.

Praktyka Teoretyczna, 2015
Crisis of Law of Value and the Real Subsumption of Labor under Capital. Questions for Michael Hardt,
Krystian Szadkowski
doaj   +1 more source

The Noisy Power Method: A Meta Algorithm with Applications [PDF]

, 2014
We provide a new robust convergence analysis of the well-known power method for computing the dominant singular vectors of a matrix that we call the noisy power method. Our result characterizes the convergence behavior of the algorithm when a significant
Hardt, Moritz, Price, Eric
core   +2 more sources

No measure for culture? Value in the new economy [PDF]

, 2010
This paper explores articulations of the value of investment in culture and the arts through a critical discourse analysis of policy documents, reports and academic commentary since 1997.
Adorno, T. W, Arvidsson, A., Belfiore, E., Belfiore, E., Bourdieu, P., Bourdieu, P., Brown, A., Caffentzis, G., Dalla Costa, M., De Angelis, M., De Angelis, M., Dowling, E., Dyer-Witheford, N., Dyer-Witheford, N., Edwards, R., Elson, D., Florida, R., Florida, R., Hardt, M., Hardt, M., Hardt, M., Hesmondhalgh, D., Holden, J., House of Commons, Hyland, M., Jowell, T, Kelly, O., Lazzarato, M., Leadbeater, C., Local Government Association (LGA), Marx, K., Marx, K., Marx, K., Matarasso, E, Mizra, M., Negri, A., Negri, A., Negri, A., Newell, S., Oakley, A., Paltridge, B., Raymond, E., Sennett, R., Tapscott, D., Terranova, T, Wenger, E.   +45 more
core   +1 more source

Personalized Differential Privacy for Ridge Regression Under Output Perturbation

Naval Research Logistics (NRL), EarlyView.
ABSTRACT The increased application of machine learning (ML) in sensitive domains requires protecting the training data through privacy frameworks, such as differential privacy (DP). Traditional DP enforces a uniform privacy level ε$$ \varepsilon $$, which bounds the maximum privacy loss that each data point in the dataset is allowed to incur.
Krishna Acharya, Franziska Boenisch, Rakshit Naidu, Juba Ziani   +3 more
wiley   +1 more source

A Biopolítica Enquanto Potência da Vida: contrapoder e os biopoderes contemporâneos [PDF]

, 2007
A abordagem crítica sobre a obra de Antonio Negri e Michael Hardt, acerca dos conceitos de biopoder e império, é orientada, neste artigo, em particular, para o modo como Hardt e Hardt são, simultaneamente, tributários do pensamento de Michel Foucault ...
Menezes, Manuel
core  

Beating Randomized Response on Incoherent Matrices [PDF]

, 2011
Computing accurate low rank approximations of large matrices is a fundamental data mining task. In many applications however the matrix contains sensitive information about individuals.
Hardt, Moritz, Roth, Aaron
core   +1 more source