A Conformal Prediction Score that is Robust to Label Noise
2405.02648
0
0
Abstract
Conformal Prediction (CP) quantifies network uncertainty by building a small prediction set with a pre-defined probability that the correct class is within this set. In this study we tackle the problem of CP calibration based on a validation set with noisy labels. We introduce a conformal score that is robust to label noise. The noise-free conformal score is estimated using the noisy labeled data and the noise level. In the test phase the noise-free score is used to form the prediction set. We applied the proposed algorithm to several standard medical imaging classification datasets. We show that our method outperforms current methods by a large margin, in terms of the average size of the prediction set, while maintaining the required coverage.
Create account to get full access
Overview
- This paper proposes a new conformal prediction score that is robust to label noise.
- Conformal prediction is a technique that provides reliable uncertainty estimates for machine learning models, but it can be sensitive to noise in the training data.
- The authors introduce a new conformal score that is designed to be more robust to label noise, improving the reliability of the resulting prediction sets.
Plain English Explanation
The paper focuses on a machine learning technique called conformal prediction. Conformal prediction provides a way to quantify the uncertainty of a machine learning model's predictions. This is important because it allows us to know how confident we should be in the model's outputs.
However, conformal prediction can be sensitive to label noise - errors or inaccuracies in the training data. When there is label noise, the conformal prediction method may produce less reliable uncertainty estimates.
To address this issue, the researchers developed a new conformal prediction score that is designed to be more robust to label noise. This means the resulting prediction sets will be more reliable, even when the training data is not perfect.
The key idea is to modify the way the conformal prediction score is calculated, taking into account the potential for label noise. This allows the method to produce better uncertainty estimates that are less affected by noisy or incorrect labels in the training data.
Technical Explanation
The paper introduces a new conformal prediction score called the "robust conformal score" (RCS). This score is designed to be more robust to label noise compared to the standard conformal prediction score.
The standard conformal score is based on the nonconformity of a test example, which measures how different the example is from the training data. The RCS modifies this by incorporating a term that estimates the probability of the label being correct. This allows the method to downweight the contribution of potentially noisy labels, leading to more reliable prediction sets.
The authors prove theoretical results showing that the RCS maintains valid coverage (the property that the prediction sets contain the true label with the desired probability) even in the presence of label noise. They also demonstrate through experiments on several benchmark datasets that the RCS outperforms the standard conformal score in terms of prediction set efficiency (the size of the prediction sets) when there is label noise.
Critical Analysis
The paper provides a solid theoretical foundation for the proposed RCS method and demonstrates its empirical effectiveness. However, there are a few potential limitations and areas for future research:
- The analysis assumes that the label noise is independent of the features, which may not always be the case in real-world applications.
- The method requires an estimate of the label noise rate, which may not be readily available in practice.
- The experiments are limited to relatively small-scale datasets, and it would be interesting to see how the RCS performs on larger, more complex problems.
- The paper does not explore the impact of the RCS on other conformal prediction properties, such as calibration and informativeness, beyond prediction set efficiency.
Overall, the RCS seems to be a promising approach for making conformal prediction more robust to label noise, but further research is needed to fully understand its strengths, limitations, and broader applicability.
Conclusion
This paper introduces a new conformal prediction score that is designed to be more robust to label noise in the training data. By incorporating an estimate of the label noise probability into the score calculation, the proposed "robust conformal score" (RCS) can produce more reliable prediction sets even when the training data is not perfect.
The theoretical and empirical results demonstrate the effectiveness of the RCS, suggesting it could be a valuable tool for deploying machine learning models in real-world scenarios where label noise is a concern. However, the method also has some limitations that merit further investigation, such as the assumption of independent label noise and the need for an estimate of the noise rate.
Overall, this research represents an important step towards making conformal prediction more practical and useful in the face of noisy or imperfect training data, which is a common challenge in many machine learning applications.
This summary was produced with help from an AI and may contain inaccuracies - check out the links to read the original source documents!
Related Papers
Verifiably Robust Conformal Prediction
Linus Jeary, Tom Kuipers, Mehran Hosseini, Nicola Paoletti
0
0
Conformal Prediction (CP) is a popular uncertainty quantification method that provides distribution-free, statistically valid prediction sets, assuming that training and test data are exchangeable. In such a case, CP's prediction sets are guaranteed to cover the (unknown) true test output with a user-specified probability. Nevertheless, this guarantee is violated when the data is subjected to adversarial attacks, which often result in a significant loss of coverage. Recently, several approaches have been put forward to recover CP guarantees in this setting. These approaches leverage variations of randomised smoothing to produce conservative sets which account for the effect of the adversarial perturbations. They are, however, limited in that they only support $ell^2$-bounded perturbations and classification tasks. This paper introduces VRCP (Verifiably Robust Conformal Prediction), a new framework that leverages recent neural network verification methods to recover coverage guarantees under adversarial attacks. Our VRCP method is the first to support perturbations bounded by arbitrary norms including $ell^1$, $ell^2$, and $ell^infty$, as well as regression tasks. We evaluate and compare our approach on image classification tasks (CIFAR10, CIFAR100, and TinyImageNet) and regression tasks for deep reinforcement learning environments. In every case, VRCP achieves above nominal coverage and yields significantly more efficient and informative prediction regions than the SotA.
6/7/2024
🔮
Provably Robust Conformal Prediction with Improved Efficiency
Ge Yan, Yaniv Romano, Tsui-Wei Weng
0
0
Conformal prediction is a powerful tool to generate uncertainty sets with guaranteed coverage using any predictive model, under the assumption that the training and test data are i.i.d.. Recently, it has been shown that adversarial examples are able to manipulate conformal methods to construct prediction sets with invalid coverage rates, as the i.i.d. assumption is violated. To address this issue, a recent work, Randomized Smoothed Conformal Prediction (RSCP), was first proposed to certify the robustness of conformal prediction methods to adversarial noise. However, RSCP has two major limitations: (i) its robustness guarantee is flawed when used in practice and (ii) it tends to produce large uncertainty sets. To address these limitations, we first propose a novel framework called RSCP+ to provide provable robustness guarantee in evaluation, which fixes the issues in the original RSCP method. Next, we propose two novel methods, Post-Training Transformation (PTT) and Robust Conformal Training (RCT), to effectively reduce prediction set size with little computation overhead. Experimental results in CIFAR10, CIFAR100, and ImageNet suggest the baseline method only yields trivial predictions including full label set, while our methods could boost the efficiency by up to $4.36times$, $5.46times$, and $16.9times$ respectively and provide practical robustness guarantee. Our codes are available at https://github.com/Trustworthy-ML-Lab/Provably-Robust-Conformal-Prediction.
5/1/2024
Conformal Prediction for Deep Classifier via Label Ranking
Jianguo Huang, Huajun Xi, Linjun Zhang, Huaxiu Yao, Yue Qiu, Hongxin Wei
0
0
Conformal prediction is a statistical framework that generates prediction sets containing ground-truth labels with a desired coverage guarantee. The predicted probabilities produced by machine learning models are generally miscalibrated, leading to large prediction sets in conformal prediction. To address this issue, we propose a novel algorithm named $textit{Sorted Adaptive Prediction Sets}$ (SAPS), which discards all the probability values except for the maximum softmax probability. The key idea behind SAPS is to minimize the dependence of the non-conformity score on the probability values while retaining the uncertainty information. In this manner, SAPS can produce compact prediction sets and communicate instance-wise uncertainty. Extensive experiments validate that SAPS not only lessens the prediction sets but also broadly enhances the conditional coverage rate of prediction sets.
6/7/2024
🔮
Conformal Prediction for Class-wise Coverage via Augmented Label Rank Calibration
Yuanjie Shi, Subhankar Ghosh, Taha Belkhouja, Janardhan Rao Doppa, Yan Yan
0
0
Conformal prediction (CP) is an emerging uncertainty quantification framework that allows us to construct a prediction set to cover the true label with a pre-specified marginal or conditional probability. Although the valid coverage guarantee has been extensively studied for classification problems, CP often produces large prediction sets which may not be practically useful. This issue is exacerbated for the setting of class-conditional coverage on imbalanced classification tasks. This paper proposes the Rank Calibrated Class-conditional CP (RC3P) algorithm to reduce the prediction set sizes to achieve class-conditional coverage, where the valid coverage holds for each class. In contrast to the standard class-conditional CP (CCP) method that uniformly thresholds the class-wise conformity score for each class, the augmented label rank calibration step allows RC3P to selectively iterate this class-wise thresholding subroutine only for a subset of classes whose class-wise top-k error is small. We prove that agnostic to the classifier and data distribution, RC3P achieves class-wise coverage. We also show that RC3P reduces the size of prediction sets compared to the CCP method. Comprehensive experiments on multiple real-world datasets demonstrate that RC3P achieves class-wise coverage and 26.25% reduction in prediction set sizes on average.
6/12/2024