output
20142021
most citedBeyond temperature scaling: Obtaining well-calibrated multiclass probabilities with Dirichlet calibration

162 citations

5 papers

stat.ME2021★ 18 cited

Modeling repeated self-reported outcome data: a continuous-time longitudinal Item Response Theory model

Cécile Proust-Lima, Viviane Philipps, Bastien Perrot +2

Item Response Theory (IRT) models have received growing interest in health science for analyzing latent constructs such as depression, anxiety, quality of life, or cognitive functi…

math.HO2020

Early-Dynastic tables from Southern Mesopotamia, or the multiple facets of the quantification of surfaces

Christine Proust

How were surfaces evaluated before the invention of the sexagesimal place value notation in Mesopotamia? This chapter examines a group of five tablets containing tables for surface…

cs.LG2019★ 162 cited

Beyond temperature scaling: Obtaining well-calibrated multiclass probabilities with Dirichlet calibration

Meelis Kull, Miquel Perello-Nieto, Markus Kängsepp +3

Class probabilities predicted by most multiclass classifiers are uncalibrated, often tending towards over-confidence. With neural networks, calibration can be improved by temperatu…

cs.CV2016★ 2 cited

Calorie Counter: RGB-Depth Visual Estimation of Energy Expenditure at Home

Lili Tao, Tilo Burghardt, Majid Mirmehdi +6

We present a new framework for vision-based estimation of calorific expenditure from RGB-D data - the first that is validated on physical gas exchange measurements and applied to d…

cs.HC2014★ 1 cited

Sensors for healthcare: Would you want them in your home?

Alison Burrows, Rachel Gooberman-Hill, Ian Craddock +1

This paper describes some of the challenges set within SPHERE, a large-scale Interdisciplinary Research Collaboration that aims to develop sensor systems to monitor people's health…