v230: Proceedings of COPA 2024

2026-06-10 · Source: Proceedings of Machine Learning Research · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Data Science & Analytics · Depth: Expert, medium

Summary

Volume 230 presents the proceedings of the 13th Symposium on Conformal and Probabilistic Prediction with Applications, held in September 2024 at Politecnico di Milano, featuring research from leading experts. The papers explore advancements in Conformal Prediction, including methods for handling data contamination, enhancing predictions with E-Test Statistics, and ensuring distribution-free risk assessment for regression algorithms. Applications span diverse fields such as binary image classification, object detection in aerial and satellite images, and multi-class classification with reject options. Further contributions address uncertainty quantification for metamodels, adaptive conformal inference for multi-step time-series forecasting, and practical tools like "crepes" for Python. The symposium also delves into critical areas like fairness considerations, anomaly detection, and the application of Conformal Prediction in active learning for predictive maintenance and housing market analysis.

Key takeaway

The 13th Symposium on Conformal and Probabilistic Prediction (CPP) presents over 30 papers advancing robust uncertainty quantification in machine learning. Key contributions include novel CPP methods for data contamination, covariate shift, and multi-step time series forecasting, alongside applications in image classification, object detection, and LLM calibration. This collection offers ML researchers and practitioners critical tools for building reliable, trustworthy AI systems with quantifiable prediction guarantees.

Topics

• Conformal Prediction
• Probabilistic Prediction
• Uncertainty Quantification
• Machine Learning Applications
• Time Series Forecasting

Best for: Research Scientist, AI Scientist, Machine Learning Engineer, Data Scientist

Related on AIssential

• v266: Proceedings of COPA 2025 — Conformal Prediction provides statistically rigorous uncertainty quantification across diverse machine learning applications and challenges.
• v179 — Volume 179 compiles the proceedings from the Eleventh Symposium on Conformal and Probabilistic Prediction with Applications, held in Brighton, UK, from August 24-26, 2022.
• Conformal Prediction with Time-Series Data via Sequential Conformalized Density Regions — SCDR offers doubly robust, asymptotically valid conformal prediction for time-series, generating informative, potentially disconnected prediction sets.
• Beyond Surface Statistics: Robust Conformal Prediction for LLMs via Internal Representations — Internal LLM representations offer more robust uncertainty signals for conformal prediction, especially under distribution shifts.
• [D] Conformal Prediction vs naive thresholding to represent uncertainty — Conformal Prediction offers theoretical guarantees for uncertainty quantification, unlike naive thresholding.
• The ninety-percent confidence that means nothing — and what does — Softmax outputs are not true probabilities and misrepresent classifier confidence; conformal prediction offers reliable coverage guarantees.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Proceedings of Machine Learning Research.