Robust Probabilistic Calibration
نویسنده
چکیده
Probabilistic calibration is the task of producing reliable estimates of the conditional class probability P (class|observation) from the outputs of numerical classifiers. A recent comparative study [1] revealed that Isotonic Regression [2] and Platt Calibration [3] are most effective probabilistic calibration technique for a wide range of classifiers. This paper will demonstrate that these methods are sensitive to outliers in the data. An improved calibration method will be introduced that combines probabilistic calibration with methods from the field of robust statistics [4]. It will be shown that the integration of robustness concepts can significantly improve calibration performance.
منابع مشابه
Robust optimal multi-objective controller design for vehicle rollover prevention
Robust control design of vehicles addresses the effect of uncertainties on the vehicle’s performance. In present study, the robust optimal multi-objective controller design on a non-linear full vehicle dynamic model with 8-degrees of freedom having parameter with probabilistic uncertainty considering two simultaneous conflicting objective functions has been made to prevent the rollover. The obj...
متن کاملRobust Probabilistic Multivariate Calibration Model
In this paper we propose a robust probabilistic multivariate calibration (RPMC) model in an attempt to identify linear relationships between two sets of observed variables contaminated with outliers. Instead of the Gaussian assumptions that predominate in classical statistical models, RPMC is closely related with the multivariate Student's t-distribution over noises and latent variables. As a r...
متن کاملCalibration of Load and Resistance Factors for Reinforced Concrete
Current approach for designing of reinforced concrete members is based on the load and resistance factor. However the load and resistance parameters are random variables, the constant values have been designated for them in the designing procedure. Assuming these factors as the constants, will be led to the unsafe and uneconomical designs. Safe designing of structures requires appropriate recog...
متن کاملRobust Depth Image Acquisition Using Modulated Pattern Projection and Probabilistic Graphical Models
Depth image acquisition with structured light approaches in outdoor environments is a challenging problem due to external factors, such as ambient sunlight, which commonly affect the acquisition procedure. This paper presents a novel structured light sensor designed specifically for operation in outdoor environments. The sensor exploits a modulated sequence of structured light projected onto th...
متن کاملLarge-scale probabilistic prediction with and without validity guarantees
This paper studies theoretically and empirically a method of turning machinelearning algorithms into probabilistic predictors that automatically enjoys a property of validity (perfect calibration) and is computationally e cient. The price to pay for perfect calibration is that these probabilistic predictors produce imprecise (in practice, almost precise for large data sets) probabilities. When ...
متن کامل