Learning Optimal Conformal Classifiers
- URL: http://arxiv.org/abs/2110.09192v1
- Date: Mon, 18 Oct 2021 11:25:33 GMT
- Title: Learning Optimal Conformal Classifiers
- Authors: David Stutz, Krishnamurthy (Dj) Dvijotham, Ali Taylan Cemgil, Arnaud
Doucet
- Abstract summary: Conformal prediction (CP) is used to predict confidence sets containing the true class with a user-specified probability.
This paper explores strategies to differentiate through CP during training with the goal of training model with the conformal wrapper end-to-end.
We show that conformal training (ConfTr) outperforms state-of-the-art CP methods for classification by reducing the average confidence set size.
- Score: 32.68483191509137
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: Modern deep learning based classifiers show very high accuracy on test data
but this does not provide sufficient guarantees for safe deployment, especially
in high-stake AI applications such as medical diagnosis. Usually, predictions
are obtained without a reliable uncertainty estimate or a formal guarantee.
Conformal prediction (CP) addresses these issues by using the classifier's
probability estimates to predict confidence sets containing the true class with
a user-specified probability. However, using CP as a separate processing step
after training prevents the underlying model from adapting to the prediction of
confidence sets. Thus, this paper explores strategies to differentiate through
CP during training with the goal of training model with the conformal wrapper
end-to-end. In our approach, conformal training (ConfTr), we specifically
"simulate" conformalization on mini-batches during training. We show that CT
outperforms state-of-the-art CP methods for classification by reducing the
average confidence set size (inefficiency). Moreover, it allows to "shape" the
confidence sets predicted at test time, which is difficult for standard CP. On
experiments with several datasets, we show ConfTr can influence how
inefficiency is distributed across classes, or guide the composition of
confidence sets in terms of the included classes, while retaining the
guarantees offered by CP.
Related papers
- Conformal Risk Minimization with Variance Reduction [37.74931189657469]
Conformal prediction (CP) is a distribution-free framework for achieving probabilistic guarantees on black-box models.
Recent research efforts have focused on optimizing CP efficiency during training.
We formalize this concept as the problem of conformal risk minimization.
arXiv Detail & Related papers (2024-11-03T21:48:15Z) - Provably Reliable Conformal Prediction Sets in the Presence of Data Poisoning [53.42244686183879]
Conformal prediction provides model-agnostic and distribution-free uncertainty quantification.
Yet, conformal prediction is not reliable under poisoning attacks where adversaries manipulate both training and calibration data.
We propose reliable prediction sets (RPS): the first efficient method for constructing conformal prediction sets with provable reliability guarantees under poisoning.
arXiv Detail & Related papers (2024-10-13T15:37:11Z) - Trustworthy Classification through Rank-Based Conformal Prediction Sets [9.559062601251464]
We propose a novel conformal prediction method that employs a rank-based score function suitable for classification models.
Our approach constructs prediction sets that achieve the desired coverage rate while managing their size.
Our contributions include a novel conformal prediction method, theoretical analysis, and empirical evaluation.
arXiv Detail & Related papers (2024-07-05T10:43:41Z) - Evidential Uncertainty Sets in Deep Classifiers Using Conformal Prediction [1.2430809884830318]
We propose Evidential Conformal Prediction (ECP) method for image classifiers to generate conformal prediction sets.
Our method is based on a non-conformity score function that has its roots in Evidential Deep Learning (EDL)
arXiv Detail & Related papers (2024-06-16T03:00:16Z) - Efficient Conformal Prediction under Data Heterogeneity [79.35418041861327]
Conformal Prediction (CP) stands out as a robust framework for uncertainty quantification.
Existing approaches for tackling non-exchangeability lead to methods that are not computable beyond the simplest examples.
This work introduces a new efficient approach to CP that produces provably valid confidence sets for fairly general non-exchangeable data distributions.
arXiv Detail & Related papers (2023-12-25T20:02:51Z) - Federated Conformal Predictors for Distributed Uncertainty
Quantification [83.50609351513886]
Conformal prediction is emerging as a popular paradigm for providing rigorous uncertainty quantification in machine learning.
In this paper, we extend conformal prediction to the federated learning setting.
We propose a weaker notion of partial exchangeability, better suited to the FL setting, and use it to develop the Federated Conformal Prediction framework.
arXiv Detail & Related papers (2023-05-27T19:57:27Z) - Fast Entropy-Based Methods of Word-Level Confidence Estimation for
End-To-End Automatic Speech Recognition [86.21889574126878]
We show how per-frame entropy values can be normalized and aggregated to obtain a confidence measure per unit and per word.
We evaluate the proposed confidence measures on LibriSpeech test sets, and show that they are up to 2 and 4 times better than confidence estimation based on the maximum per-frame probability.
arXiv Detail & Related papers (2022-12-16T20:27:40Z) - Confidence Estimation via Auxiliary Models [47.08749569008467]
We introduce a novel target criterion for model confidence, namely the true class probability ( TCP)
We show that TCP offers better properties for confidence estimation than standard maximum class probability (MCP)
arXiv Detail & Related papers (2020-12-11T17:21:12Z) - Failure Prediction by Confidence Estimation of Uncertainty-Aware
Dirichlet Networks [6.700873164609009]
It is shown that uncertainty-aware deep Dirichlet neural networks provide an improved separation between the confidence of correct and incorrect predictions in the true class probability metric.
A new criterion is proposed for learning the true class probability by matching prediction confidence scores while taking imbalance and TCP constraints into account.
arXiv Detail & Related papers (2020-10-19T21:06:45Z) - Binary Classification from Positive Data with Skewed Confidence [85.18941440826309]
Positive-confidence (Pconf) classification is a promising weakly-supervised learning method.
In practice, the confidence may be skewed by bias arising in an annotation process.
We introduce the parameterized model of the skewed confidence, and propose the method for selecting the hyper parameter.
arXiv Detail & Related papers (2020-01-29T00:04:36Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.