New research enhances conformal prediction for fairness and efficiency
ByPulseAugur Editorial·[15 sources]·
Researchers are advancing conformal prediction (CP) techniques to improve uncertainty quantification and fairness in machine learning. New methods like FedCF aim to extend CP to federated learning settings, enabling fairness audits across different subgroups. Other advancements include DistMatch for robust sequential CP in time series, SpeedCP for efficient kernel-based conditional CP, and DCO for decoupled optimization of prediction sets. Additionally, new diagnostics like ERT are being developed to better evaluate conditional coverage, and research is exploring substantive fairness beyond procedural guarantees.
AI
IMPACT
These advancements in conformal prediction offer improved methods for uncertainty quantification, fairness, and robustness, crucial for reliable AI deployment in sensitive applications.
RANK_REASON
The cluster consists of multiple academic papers detailing new methods and analyses within the field of conformal prediction.
arXiv:2509.22907v2 Announce Type: replace Abstract: Conformal Prediction (CP) is a widely used technique for quantifying uncertainty in machine learning models. In its standard form, CP offers probabilistic guarantees on the coverage of the true label, but it is agnostic to sensi…
arXiv:2606.00690v1 Announce Type: new Abstract: Sequential conformal prediction (CP) provides valid uncertainty quantification under the assumption of residual exchangeability. However, this assumption is often violated in real-world time series due to temporal dependencies and d…
arXiv cs.LG
TIER_1English(EN)·Yating Liu, Yeo Jin Jung, Zixuan Wu, So Won Jeong, Claire Donnat·
arXiv:2509.24100v2 Announce Type: replace-cross Abstract: Conformal prediction provides distribution-free prediction sets with finite-sample conditional guarantees. We build upon the RKHS-based framework of Gibbs et al. (2023), which leverages families of covariate shifts to prov…
arXiv:2605.24983v1 Announce Type: new Abstract: Conformal prediction is a useful and versatile alternative to model calibration in machine learning classification. It replaces single-class prediction with prediction sets, guaranteeing that the \textit{a priori} probability of the…
Bayesian conformal optimisation methods often use the same held-out data both to search for efficient prediction sets and to certify coverage or risk. This coupling is natural for high-probability risk-control guarantees, but it is not necessary when the target is standard finite…
arXiv:2606.00419v1 Announce Type: new Abstract: Uncertainty quantification (UQ) is critical for the deployment of machine learning predictors in real-world scenarios where the data distribution may shift over time (i.e., data may not be exchangeable). Online conformal prediction …
arXiv stat.ML
TIER_1English(EN)·Martin V. Vejling, Christophe A. N. Biscio, Adrien Mazoyer, Petar Popovski, Shashi Raj Pandey·
arXiv:2606.00717v1 Announce Type: cross Abstract: Uncertainty quantification is essential in high-stakes machine learning tasks. However, one of the principled solutions, conformal prediction, faces challenges under limited local calibration data, privacy constraints, and data he…
arXiv stat.ML
TIER_1English(EN)·Pengqi Liu, Zijun Yu, Mouloud Belbahri, Arthur Charpentier, Masoud Asgharian, Jesse C. Cresswell·
arXiv:2509.04631v2 Announce Type: replace-cross Abstract: Transductive conformal prediction addresses the simultaneous prediction for multiple data points. Given a desired confidence level, the objective is to construct a prediction set that includes the true outcomes with the pr…
arXiv stat.ML
TIER_1English(EN)·Sacha Braun, David Holzm\"uller, Michael I. Jordan, Francis Bach·
arXiv:2512.11779v2 Announce Type: replace Abstract: Evaluating conditional coverage remains one of the most persistent challenges in assessing the reliability of predictive systems. Although conformal methods can give guarantees on marginal coverage, no method can guarantee to pr…
Uncertainty quantification is essential in high-stakes machine learning tasks. However, one of the principled solutions, conformal prediction, faces challenges under limited local calibration data, privacy constraints, and data heterogeneity. In multi-agent settings, existing wor…
Uncertainty quantification (UQ) is critical for the deployment of machine learning predictors in real-world scenarios where the data distribution may shift over time (i.e., data may not be exchangeable). Online conformal prediction (OCP) methods address this issue at the expense …
arXiv:2411.13479v4 Announce Type: replace Abstract: We consider conformal prediction for multivariate data and focus on hierarchical data, where some components are linear combinations of others. Intuitively, the hierarchical structure can be leveraged to reduce the size of predi…
arXiv stat.ML
TIER_1English(EN)·Yao Zhang, Emmanuel J. Cand\`es·
arXiv:2409.19712v2 Announce Type: replace-cross Abstract: Conformal prediction is a popular technique for constructing prediction intervals with distribution-free coverage guarantees. The coverage is marginal, meaning it only holds on average over the entire population but not ne…
arXiv:2510.04406v2 Announce Type: replace Abstract: Conformal prediction offers finite-sample coverage guarantees under minimal assumptions. However, existing methods treat the entire modeling process as a black box, overlooking opportunities to exploit and understand modular str…
