Constructing Normalized Nonconformity Measures based on Maximizing Predictive Efficiency

Research output: Journal PublicationConference articlepeer-review

3 Citations (Scopus)

Abstract

The problem of regression in the inductive conformal prediction framework is addressed to provide prediction intervals that are optimized by predictive efficiency. A differentiable function is used to approximate the exact optimization problem of minimizing predictive inefficiency on a training data set using a conformal predictor based on a parametric normalized nonconformity measure. Gradient descent is then used to find a solution. Since the optimization approximates the conformal predictor, this method is called surrogate conformal predictor optimization. Experiments are reported that show that it results in conformal predictors that provide improved predictive efficiency for regression problems on several data sets, whilst remaining reliable. It is also shown that the optimal parameter values typically differ for different confidence levels. Using house price data, alternative measures of inefficiency are explored to address different application requirements.

Original languageEnglish
Pages (from-to)41-54
Number of pages14
JournalProceedings of Machine Learning Research
Volume128
Publication statusPublished - 2020
Event9th Symposium on Conformal and Probabilistic Predictions with Applications, COPA 2020 - Virtual, Online, Italy
Duration: 9 Sept 202011 Sept 2020

Keywords

  • Conformal prediction
  • gradient descent
  • regression

ASJC Scopus subject areas

  • Artificial Intelligence
  • Software
  • Control and Systems Engineering
  • Statistics and Probability

Fingerprint

Dive into the research topics of 'Constructing Normalized Nonconformity Measures based on Maximizing Predictive Efficiency'. Together they form a unique fingerprint.

Cite this