[CFP] PRJ Special Issue on Conformal Prediction and Distribution-Free Uncertainty Quantification

[Harris Papadopoulos]

[ Please distribute - apologies for multiple postings ]

*** Call for Papers ***

Pattern Recognition Journal Special Issue on

Conformal Prediction and Distribution-Free Uncertainty Quantification

Details: https://www.sciencedirect.com/journal/pattern-recognition/about/call-for-papers

Important Dates

• Submissions accepted until: September 30, 2024
• Final notifications before: May 31, 2025

The review process will commence upon each submission and notifications will be communicated accordingly.

Submission Information

Submission link: https://www.editorialmanager.com/PR/

Select article type VSI: Conformal Prediction

Theme

In the rapidly evolving landscape of Machine Learning and Pattern Recognition, the emergence and development of Conformal Prediction (CP) have marked a significant stride towards more reliable predictive models. CP stands out as a versatile framework capable of offering prediction regions with valid coverage guarantees under minimal assumptions. Its unique strength lies in providing probabilistically valid guarantees, which are of paramount importance for complex Pattern Recognition tasks in computer vision, biometrics, text analysis, and other areas where decision certainty is crucial. Highlighting its continual evolution, recent advancements such as Conformal Predictive Distributions, Conformal change-point detection, Venn-Abers predictors, and enhancements like jackknife+ have opened new frontiers in CP, enhancing its applicability and effectiveness.

The relevance and growing interest in CP are evident from its increasing presence in academic literature and conferences. In the past few years, the field has seen a consistent growth in scholarly attention, highlighted by the annual Symposium on Conformal and Probabilistic Predictions with Applications (COPA) since 2012, workshops at ICML in 2021 and 2022, and several special issues in leading journals. These platforms have not only contributed to the dissemination of knowledge, but also to the advancement of CP as a critical tool in the realm of Pattern Recognition.

This special issue aims to explore innovative research in the rapidly evolving field of Conformal Prediction (CP), focusing on its integration and application within the broader scope of Pattern Recognition. We welcome submissions that demonstrate novel approaches, theoretical insights, and practical applications of CP and its role in advancing the field of Pattern Recognition.

Topics of interest include, but are not limited to:

• Theoretical analyses and performance guarantees of CP.
• Novel CP approaches and conformity measures.
• Conformal predictive distributions.
• Conformal change-point and anomaly detection.
• Venn-Abers and other multiprobability prediction approaches.
• Post-hoc calibration through CP.
• Decision-making through CP and distribution-free uncertainty quantification.
• Implementations of CP frameworks and algorithms.
• CP for explainable machine learning and Fairness, Accountability and Transparency (FAT).
• CP applications in computer vision, image processing, biometrics, security, text analysis and more.
• CP in real-world applications: challenges and solutions.

Guest Editors

Harris Papadopoulos, Frederick University, Nicosia, Cyprus

Khuong An Nguyen, Royal Holloway University of London, Egham, UK

Vineeth Balasubramanian, Indian Institute of Technology, Hyderabad, India