However, the design of methods for learning credal set predictors remains a challenging problem. In this paper, we make use of conformal prediction for this purpose.

Uncertainty Quantification (UQ) for Natural Language Generation (NLG) is crucial for assessing the performance of Large Language Models (LLMs), as it reveals confidence in predictions, identifies failure modes, and gauges output reliability. Conformal Prediction (CP), a model-agnostic method that generates prediction sets with a specified error rate, has been adopted for UQ in classification tasks, where the size of the prediction set indicates the model's uncertainty. However, when adapting CP to NLG, the sampling-based method for generating candidate outputs cannot guarantee the inclusion of the ground truth, limiting its applicability across a wide range of error rates. To address this, we propose \ourmethod, a method that explicitly adds the ground truth to the candidate outputs and uses logit scores to measure nonconformity. Our experiments with six LLMs on four NLG tasks show that \ourmethod outperforms baseline methods in calibrating error rates and empirical cover rates, offering accurate UQ across a wide range of user-specified error rates.

Supervised machine learning techniques have shown promising results in code analysis and optimization problems. However, a learning-based solution can be brittle because minor changes in hardware or application workloads -- such as facing a new CPU architecture or code pattern -- may jeopardize decision accuracy, ultimately undermining model robustness. We introduce Prom, an open-source library to enhance the robustness and performance of predictive models against such changes during deployment. Prom achieves this by using statistical assessments to identify test samples prone to mispredictions and using feedback on these samples to improve a deployed model. We showcase Prom by applying it to 13 representative machine learning models across 5 code analysis and optimization tasks. Our extensive evaluation demonstrates that Prom can successfully identify an average of 96% (up to 100%) of mispredictions. By relabeling up to 5% of the Prom-identified samples through incremental learning, Prom can help a deployed model achieve a performance comparable to that attained during its model training phase.

Credal sets are sets of probability distributions that are considered as candidates for an imprecisely known ground-truth distribution. In machine learning, they have recently attracted attention as an appealing formalism for uncertainty representation, in particular due to their ability to represent both the aleatoric and epistemic uncertainty in a prediction. However, the design of methods for learning credal set predictors remains a challenging problem. In this paper, we make use of conformal prediction for this purpose. More specifically, we propose a method for predicting credal sets in the classification task, given training data labeled by probability distributions. Since our method inherits the coverage guarantees of conformal prediction, our conformal credal sets are guaranteed to be valid with high probability (without any assumptions on model or distribution). We demonstrate the applicability of our method to natural language inference, a highly ambiguous natural language task where it is common to obtain multiple annotations per example.

An important mathematical tool in the analysis of dynamical systems is the approximation of the reach set, i.e., the set of states reachable after a given time from a given initial state. This set is difficult to compute for complex systems even if the system dynamics are known and given by a system of ordinary differential equations with known coefficients. In practice, parameters are often unknown and mathematical models difficult to obtain. Data-based approaches are promised to avoid these difficulties by estimating the reach set based on a sample of states. If a model is available, this training set can be obtained through numerical simulation. In the absence of a model, real-life observations can be used instead. A recently proposed approach for data-based reach set approximation uses Christoffel functions to approximate the reach set. Under certain assumptions, the approximation is guaranteed to converge to the true solution. In this paper, we improve upon these results by notably improving the sample efficiency and relaxing some of the assumptions by exploiting statistical guarantees from conformal prediction with training and calibration sets. In addition, we exploit an incremental way to compute the Christoffel function to avoid the calibration set while maintaining the statistical convergence guarantees. Furthermore, our approach is robust to outliers in the training and calibration set.

The property of conformal predictors to guarantee the required accuracy rate makes this framework attractive in various practical applications. However, this property is achieved at a price of reduction in precision. In the case of conformal classification, the systems can output multiple class labels instead of one. It is also known from the literature, that the choice of nonconformity function has a major impact on the efficiency of conformal classifiers. Recently, it was shown that different model-agnostic nonconformity functions result in conformal classifiers with different characteristics. For a Neural Network-based conformal classifier, the inverse probability (or hinge loss) allows minimizing the average number of predicted labels, and margin results in a larger fraction of singleton predictions. In this work, we aim to further extend this study. We perform an experimental evaluation using 8 different classification algorithms and discuss when the previously observed relationship holds or not. Additionally, we propose a successful method to combine the properties of these two nonconformity functions. The experimental evaluation is done using 11 real and 5 synthetic datasets.

Cyber-physical systems (CPS) can benefit by the use of learning enabled components (LECs) such as deep neural networks (DNNs) for perception and desicion making tasks. However, DNNs are typically non-transparent making reasoning about their predictions very difficult, and hence their application to safety-critical systems is very challenging. LECs could be integrated easier into CPS if their predictions could be complemented with a confidence measure that quantifies how much we trust their output. The paper presents an approach for computing confidence bounds based on Inductive Conformal Prediction (ICP). We train a Triplet Network architecture to learn representations of the input data that can be used to estimate the similarity between test examples and examples in the training data set. Then, these representations are used to estimate the confidence of a set predictor based on the DNN architecture used in the triplet based on ICP. The approach is evaluated using a robotic navigation benchmark and the results show that we can computed trusted confidence bounds efficiently in real-time.

Machine learning components such as deep neural networks are used extensively in Cyber-physical Systems (CPS). However, they may introduce new types of hazards that can have disastrous consequences and need to be addressed for engineering trustworthy systems. Although deep neural networks offer advanced capabilities, they must be complemented by engineering methods and practices that allow effective integration in CPS. In this paper, we investigate how to use the conformal prediction framework for assurance monitoring of CPS with machine learning components. In order to handle high-dimensional inputs in real-time, we compute nonconformity scores using embedding representations of the learned models. By leveraging conformal prediction the approach provides well-calibrated confidence and can allow monitoring that ensures a bounded small error rate while limiting the number of inputs for which an accurate prediction cannot be made. Empirical evaluation results using the German Traffic Sign Recognition Benchmark and a robot navigation dataset demonstrate that the error rates are well-calibrated while the number of alarms is small. The method is computationally efficient, and therefore, the approach is promising for assurance monitoring of CPS.