2,937 research outputs found
Writers Talk Featuring Sonya Huber
No full textSonya Huber, 2004 graduate of OSU's MFA Creative Writing Program, currently an assistant professor at Georgia Southern University. Author of "The Backwards Research Guide for Writers," "Opa Nobody," and most recently "Cover Me: A Health Insurance Memoir."The media can be accessed here: http://streaming.osu.edu/knowledgebank/cstw12/WT_WCRS_11-08-10_SonyaHuber.mp3Ohio State University. Center for the Study and Teaching of Writin
Experimental Data and Models for "Knowledge-Based Modeling of Simulation Behavior for Bayesian Optimization"
No full textThese files contain the data for the OpenDiHu experiments in sections 5.1.3 and 5.2.
The settings used for the OpenDiHu simulations are in the opendihu/ folder.
The simulation data is in the data/ folder and the used stan models are in the models/ folder.
For more details see README.md
[Hyponatraemia in patients with neurosurgical disorders: SIADH or cerebral salt wasting syndrome?]
No full textPatients with neurosurgical disorders often present with hyponatraemia. Two mechanisms account for hyponatraemia in these patients: the Syndrome of Inappropriate Secretion of Antidiuretic Hormone (SIADH) and Cerebral Salt Wasting Syndrome (CSWS). The two entities differ in their volume status. In SIADH, volume is expanded due to ADH-mediated renal water retention, but in CSWS, volume is diminished as a consequence of renal salt wasting, most likely attributable to an increased secretion of Brain Natriuretic Peptide (BNP) and Artrial Natriuretic Peptide (ANP). Since it is clinically difficult to distinguish between these two entities, fluid management has to be performed carefully. Salt and fluid replacement appears to be indicated in CSWS, whereas fluid restriction might be the primary approach in patients with SIADH
Robust Linear and Support Vector Regression
No full textThe robust Huber M-estimator, a differentiable cost function that is quadratic for small errors and linear otherwise, is
modeled exactly, in the original primal space of the problem, by an easily solvable simple convex quadratic program for both linear and
nonlinear support vector estimators. Previous models were significantly more complex or formulated in the dual space and most
involved specialized numerical algorithms for solving the robust Huber linear estimator [3], [6], [12], [13], [14], [23], [28]. Numerical test
comparisons with these algorithms indicate the computational effectiveness of the new quadratic programming model for both linear
and nonlinear support vector problems. Results are shown on problems with as many as 20,000 data points, with considerably faster
running times on larger problems
- …
