Skip to main content
SpringerLink
Log in
Springer
Machine Learning
Publishing model:
Hybrid

Machine Learning - Call for Papers: Conformal Prediction and Distribution-Free Uncertainty Quantification

Conformal prediction is a framework that allows for controlling the error rate of any predictive model. This is achieved by turning point predictions into set predictions, which are guaranteed to include the true target with a probability that is specified by the user. Various related frameworks and algorithms have been developed, such as conformal predictive systems, Venn predictors, and jackknife+. The aim of this special issue is to present novel contributions on conformal prediction and related frameworks that provide validity guarantees with minimal distributional and model assumptions.

The growing popularity of conformal prediction and distribution-free uncertainty quantification is reflected in an increasing number of papers in academic journals and conference proceedings. Since 2012, there has been an annual Symposium on Conformal and Probabilistic Predictions with Applications (COPA). In conjunction with ICML 2021 and 2022, two workshops were held on distribution-free uncertainty quantification. There were also a previous special issue of Machine Learning journal in 2018, two special issues of the Annals of Mathematics and Artificial Intelligence in 2015 and 2017, and a special issue of the Pattern Recognition journal in 2022 devoted to conformal and probabilistic prediction with applications.

The main aim of this special issue is to update the readers with the latest developments in this field and outline several new directions for future research. It is widely open for contributions concerning the theory and practice of conformal prediction and distribution-free uncertainty quantification.

Topics of interest include, but are not limited to:

  • Theoretical analyses, including performance guarantees
  • Novel conformity measures
  • Conformal change-point and anomaly detection
  • Venn-Abers predictors and other approaches to multiprobability prediction
  • Post-hoc calibration
  • Conformal predictive distributions
  • Probabilistic prediction
  • Decision-making using conformal prediction and distribution-free uncertainty quantification
  • Implementations of frameworks and algorithms
  • Applications to explainable machine learning and Fairness, Accountability and Transparency (FAT) and within domains, such as bioinformatics, drug discovery, medicine, and information security

Schedule

Submission Deadline: Jan. 7, 2024
Review Deadline: March 31, 2024 
Revision Deadline: April 30, 2024
Camera-ready Manuscript Deadline: June 30, 2024


Guest Editors

Prof. Henrik Boström, KTH Royal Institute of Technology, Sweden; bostromh@kth.se
Prof. Eyke Hüllermeier, Ludwig-Maximilians-Universität München, Germany; eyke@ifi.lmu.de
Prof. Ulf Johansson, Jönköping University, Sweden; ulf.johansson@ju.se
Dr. Khuong An Nguyen, University of Brighton, UK; K.A.Nguyen@brighton.ac.uk
Ass. Prof. Aaditya Ramdas, Carnegie Mellon University; aramdas@cs.cmu.edu

Search

Navigation