Single-cell reference atlases are large-scale, cell-level maps that capture cellular heterogeneity within an organ using single cell genomics. Given their size and cellular diversity, these atlases serve as high-quality training data for the transfer of cell type labels to new datasets. Such label transfer, however, must be robust to domain shifts in gene expression due to measurement technique, lab specifics and more general batch effects. This requires methods that provide uncertainty estimates on the cell type predictions to ensure correct interpretation. Here, for the first time, we introduce uncertainty quantification methods for cell type classification on single-cell reference atlases. We benchmark four model classes and show that currently used models lack calibration, robustness, and actionable uncertainty scores. Furthermore, we demonstrate how models that quantify uncertainty are better suited to detect unseen cell types in the setting of atlas-level cell type transfer.

In the past years, deep learning has seen an increase of usage in the domain of histopathological applications. However, while these approaches have shown great potential, in high-risk environments deep learning models need to be able to judge their own uncertainty and be able to reject inputs when there is a significant chance of misclassification. In this work, we conduct a rigorous evaluation of the most commonly used uncertainty and robustness methods for the classification of Whole-Slide-Images under domain shift using the H\&E stained Camelyon17 breast cancer dataset. Although it is known that histopathological data can be subject to strong domain shift and label noise, to our knowledge this is the first work that compares the most common methods for uncertainty estimation under these aspects. In our experiments, we compare Stochastic Variational Inference, Monte-Carlo Dropout, Deep Ensembles, Test-Time Data Augmentation as well as combinations thereof. We observe that ensembles of methods generally lead to higher accuracies and better calibration and that Test-Time Data Augmentation can be a promising alternative when choosing an appropriate set of augmentations. Across methods, a rejection of the most uncertain tiles leads to a significant increase in classification accuracy on both in-distribution as well as out-of-distribution data. Furthermore, we conduct experiments comparing these methods under varying conditions of label noise. We observe that the border regions of the Camelyon17 dataset are subject to label noise and evaluate the robustness of the included methods against different noise levels. Lastly, we publish our code framework to facilitate further research on uncertainty estimation on histopathological data.

深度神经网络具有令人印象深刻的性能，但是他们无法可靠地估计其预测信心，从而限制了其在高风险领域中的适用性。我们表明，应用多标签的一VS损失揭示了分类的歧义并降低了模型的过度自信。引入的Slova（单标签One-Vs-All）模型重新定义了单个标签情况的典型单VS-ALL预测概率，其中只有一个类是正确的答案。仅当单个类具有很高的概率并且其他概率可忽略不计时，提议的分类器才有信心。与典型的SoftMax函数不同，如果所有其他类的概率都很小，Slova自然会检测到分布的样本。该模型还通过指数校准进行了微调，这使我们能够与模型精度准确地对齐置信分数。我们在三个任务上验证我们的方法。首先，我们证明了斯洛伐克与最先进的分布校准具有竞争力。其次，在数据集偏移下，斯洛伐克的性能很强。最后，我们的方法在检测到分布样品的检测方面表现出色。因此，斯洛伐克是一种工具，可以在需要不确定性建模的各种应用中使用。

人工智能的最新趋势是将验证的模型用于语言和视觉任务，这些模型已经实现了非凡的表现，但也令人困惑。因此，以各种方式探索这些模型的能力对该领域至关重要。在本文中，我们探讨了模型的可靠性，在其中我们将可靠的模型定义为一个不仅可以实现强大的预测性能，而且在许多涉及不确定性（例如选择性预测，开放式设置识别）的决策任务上，在许多决策任务上表现出色，而且表现良好。强大的概括（例如，准确性和适当的评分规则，例如在分布数据集中和分发数据集上的对数可能性）和适应性（例如，主动学习，几乎没有射击不确定性）。我们设计了40个数据集的10种任务类型，以评估视觉和语言域上可靠性的不同方面。为了提高可靠性，我们分别开发了VIT-PLEX和T5-PLEX，分别针对视觉和语言方式扩展了大型模型。 PLEX极大地改善了跨可靠性任务的最先进，并简化了传统协议，因为它可以改善开箱即用的性能，并且不需要设计分数或为每个任务调整模型。我们演示了高达1B参数的模型尺寸的缩放效果，并预处理数据集大小最多4B示例。我们还展示了PLEX在具有挑战性的任务上的功能，包括零射门的开放式识别，主动学习和对话语言理解中的不确定性。

如果不确定性量化（UQ）对于实现值得信赖的机器学习至关重要，则大多数UQ方法都遭受不同和不一致的评估协议。我们声称这种不一致的原因是社区对UQ的不明确要求。本意见论文通过通过五个下游任务指定这些要求来提供新的观点，我们期望不确定性得分具有实质性的预测能力。我们仔细设计了这些下游任务，以反映ML模型的现实用法。在7个分类数据集的示例基准上，我们没有观察到最新的内在UQ方法与简单基线的统计优势。我们认为，我们的发现质疑为什么我们量化不确定性并呼吁根据被证明与ML从业人员相关的指标进行标准化协议进行标准化协议。

Modern machine learning methods including deep learning have achieved great success in predictive accuracy for supervised learning tasks, but may still fall short in giving useful estimates of their predictive uncertainty. Quantifying uncertainty is especially critical in real-world settings, which often involve input distributions that are shifted from the training distribution due to a variety of factors including sample bias and non-stationarity. In such settings, well calibrated uncertainty estimates convey information about when a model's output should (or should not) be trusted. Many probabilistic deep learning methods, including Bayesian-and non-Bayesian methods, have been proposed in the literature for quantifying predictive uncertainty, but to our knowledge there has not previously been a rigorous largescale empirical comparison of these methods under dataset shift. We present a largescale benchmark of existing state-of-the-art methods on classification problems and investigate the effect of dataset shift on accuracy and calibration. We find that traditional post-hoc calibration does indeed fall short, as do several other previous methods. However, some methods that marginalize over models give surprisingly strong results across a broad spectrum of tasks.

Reliable application of machine learning-based decision systems in the wild is one of the major challenges currently investigated by the field. A large portion of established approaches aims to detect erroneous predictions by means of assigning confidence scores. This confidence may be obtained by either quantifying the model's predictive uncertainty, learning explicit scoring functions, or assessing whether the input is in line with the training distribution. Curiously, while these approaches all state to address the same eventual goal of detecting failures of a classifier upon real-life application, they currently constitute largely separated research fields with individual evaluation protocols, which either exclude a substantial part of relevant methods or ignore large parts of relevant failure sources. In this work, we systematically reveal current pitfalls caused by these inconsistencies and derive requirements for a holistic and realistic evaluation of failure detection. To demonstrate the relevance of this unified perspective, we present a large-scale empirical study for the first time enabling benchmarking confidence scoring functions w.r.t all relevant methods and failure sources. The revelation of a simple softmax response baseline as the overall best performing method underlines the drastic shortcomings of current evaluation in the abundance of publicized research on confidence scoring. Code and trained models are at https://github.com/IML-DKFZ/fd-shifts.

The ability to quickly and accurately identify covariate shift at test time is a critical and often overlooked component of safe machine learning systems deployed in high-risk domains. While methods exist for detecting when predictions should not be made on out-of-distribution test examples, identifying distributional level differences between training and test time can help determine when a model should be removed from the deployment setting and retrained. In this work, we define harmful covariate shift (HCS) as a change in distribution that may weaken the generalization of a predictive model. To detect HCS, we use the discordance between an ensemble of classifiers trained to agree on training data and disagree on test data. We derive a loss function for training this ensemble and show that the disagreement rate and entropy represent powerful discriminative statistics for HCS. Empirically, we demonstrate the ability of our method to detect harmful covariate shift with statistical certainty on a variety of high-dimensional datasets. Across numerous domains and modalities, we show state-of-the-art performance compared to existing methods, particularly when the number of observed test samples is small.

整个幻灯片组织学图像中的组织类型学注释是一项复杂而乏味但既繁琐但必要的任务，用于开发计算病理学模型。我们建议通过将开放式识别技术应用于共同分类属于一组带注释类的组织的任务来解决此问题。临床相关的组织类别，同时拒绝测试时间开放式样品，即属于训练集中不存在的类别的图像。为此，我们引入了一种基于训练模型的开放式组织病理图像识别的新方法，以准确识别图像类别，并同时预测已应用了哪些数据增强变换。在测试时间中，我们测量了模型的置信度预测这种转换，我们期望开放集中的图像较低。在组织学图像的结直肠癌评估的背景下，我们进行了全面的实验，这些实验为我们的方法提供了证据，以自动从未知类别中识别样品的优势。代码在https://github.com/agaldran/t3po上发布。

数据质量的系统量化对于一致的模型性能至关重要。先前的工作集中在分发数据上。取而代之的是，我们解决了一个研究了一个研究的且同样重要的问题，即表征不协调的区域（ID）数据，这可能是由特征空间异质性引起的。为此，我们提出了使用数据套件的范式转移：一个以数据为中心的AI框架来识别这些区域，而与特定于任务的模型无关。数据套件利用Copula建模，表示学习和共形预测，以基于一组培训实例来构建功能置信区间估计器。这些估计器可用于评估有关培训集的测试实例的一致性，以回答两个实际有用的问题：（1）通过培训培训实例培训的模型可以可靠地预测哪些测试实例？ （2）我们可以确定功能空间的不协调区域，以便数据所有者了解数据的局限性还是指导未来数据收集？我们从经验上验证了数据套件的性能和覆盖范围保证，并在跨站点的医疗数据，有偏见的数据以及具有概念漂移的数据上证明，数据套件最能确定下游模型可能是可靠的ID区域（与所述模型无关）。我们还说明了这些确定的区域如何为数据集提供见解并突出其局限性。

室外（OOD）检测是面向任务的对话框系统中的关键组件，旨在确定查询是否不在预定义的支持的意图集之外。事实证明，先前基于软磁性的检测算法对OOD样品被过度自信。在本文中，我们分析了过度自信的OOD来自由于训练和测试分布之间的不匹配而导致的分布不确定性，这使得该模型无法自信地做出预测，因此可能导致异常软磁得分。我们提出了一个贝叶斯OOD检测框架，以使用Monte-Carlo辍学来校准分布不确定性。我们的方法是灵活的，并且可以轻松地插入现有的基于软磁性的基线和增益33.33 \％OOD F1改进，而与MSP相比仅增加了0.41 \％的推理时间。进一步的分析表明，贝叶斯学习对OOD检测的有效性。

用于图形分类的分布外检测的问题远未解决。现有模型往往对OOD示例过高自信，或者完全忽略检测任务。在这项工作中，我们从不确定性估计的角度考虑了这个问题，并进行了几种最近提出的方法的比较。在我们的实验中，我们发现没有通用的OOD检测方法，并且重要的是考虑图表和预测分类分布。

飞机行业不断努力在人类的努力，计算时间和资源消耗方面寻求更有效的设计优化方法。当替代模型和最终过渡到HF模型的开关机制均被正确校准时，混合替代物优化保持了高效果，同时提供快速的设计评估。前馈神经网络（FNN）可以捕获高度非线性输入输出映射，从而为飞机绩效因素提供有效的替代物。但是，FNN通常无法概括分布（OOD）样本，这阻碍了它们在关键飞机设计优化中的采用。通过Smood，我们基于平滑度的分布检测方法，我们建议用优化的FNN替代物来编码一个依赖模型的OOD指标，以产生具有选择性但可信度的预测的值得信赖的替代模型。与常规的不确定性接地方法不同，Smood利用了HF模拟的固有平滑性特性，可以通过揭示其可疑敏感性有效地暴露OOD，从而避免对OOD样品的过度自信不确定性估计。通过使用SMOOD，仅将高风险的OOD输入转发到HF模型以进行重新评估，从而以低开销成本获得更准确的结果。研究了三个飞机性能模型。结果表明，基于FNN的代理在预测性能方面优于其高斯过程。此外，在所有研究案例中，Smood的确覆盖了85％的实际OOD。当Smood Plus FNN替代物被部署在混合替代优化设置中时，它们的错误率分别降低了34.65％和计算速度的降低率分别为58.36次。

尽管对检测分配（OOD）示例的重要性一致，但就OOD示例的正式定义几乎没有共识，以及如何最好地检测到它们。我们将这些示例分类为它们是否表现出背景换档或语义移位，并发现ood检测，模型校准和密度估计（文本语言建模）的两个主要方法，对这些类型的ood数据具有不同的行为。在14对分布和ood英语自然语言理解数据集中，我们发现密度估计方法一致地在背景移位设置中展开校准方法，同时在语义移位设置中执行更糟。此外，我们发现两种方法通常都无法检测到挑战数据中的示例，突出显示当前方法的弱点。由于没有单个方法在所有设置上都效果很好，因此在评估不同的检测方法时，我们的结果呼叫了OOD示例的明确定义。

本文我们的目标是利用异质的温度缩放作为校准策略（OOD）检测。此处的异质性是指每个样品的最佳温度参数可能不同，而不是传统的方法对整个分布使用相同的值。为了实现这一目标，我们提出了一种称为锚定的新培训策略，可以估算每个样品的适当温度值，从而导致几个基准的最新OOD检测性能。使用NTK理论，我们表明该温度函数估计与分类器的认知不确定性紧密相关，这解释了其行为。与某些表现最佳的OOD检测方法相反，我们的方法不需要暴露于其他离群数据集，自定义校准目标或模型结合。通过具有不同OOD检测设置的经验研究 - 远处，OOD附近和语义相干OOD - 我们建立了一种高效的OOD检测方法。可以在此处访问代码和模型-https：//github.com/rushilanirudh/amp

Deep learning models that leverage large datasets are often the state of the art for modelling molecular properties. When the datasets are smaller (< 2000 molecules), it is not clear that deep learning approaches are the right modelling tool. In this work we perform an extensive study of the calibration and generalizability of probabilistic machine learning models on small chemical datasets. Using different molecular representations and models, we analyse the quality of their predictions and uncertainties in a variety of tasks (binary, regression) and datasets. We also introduce two simulated experiments that evaluate their performance: (1) Bayesian optimization guided molecular design, (2) inference on out-of-distribution data via ablated cluster splits. We offer practical insights into model and feature choice for modelling small chemical datasets, a common scenario in new chemical experiments. We have packaged our analysis into the DIONYSUS repository, which is open sourced to aid in reproducibility and extension to new datasets.

Single-cell transcriptomics enabled the study of cellular heterogeneity in response to perturbations at the resolution of individual cells. However, scaling high-throughput screens (HTSs) to measure cellular responses for many drugs remains a challenge due to technical limitations and, more importantly, the cost of such multiplexed experiments. Thus, transferring information from routinely performed bulk RNA HTS is required to enrich single-cell data meaningfully. We introduce chemCPA, a new encoder-decoder architecture to study the perturbational effects of unseen drugs. We combine the model with an architecture surgery for transfer learning and demonstrate how training on existing bulk RNA HTS datasets can improve generalisation performance. Better generalisation reduces the need for extensive and costly screens at single-cell resolution. We envision that our proposed method will facilitate more efficient experiment designs through its ability to generate in-silico hypotheses, ultimately accelerating drug discovery.

不确定性量化（UQ）对于创建值得信赖的机器学习模型至关重要。近年来，UQ方法急剧上升，可以标记可疑的例子，但是，通常不清楚这些方法确切地识别出什么。在这项工作中，我们提出了一种假设轻型方法来解释UQ模型本身。我们介绍了混淆密度矩阵 - 基于内核的错误分类密度的近似 - 并使用它将给定UQ方法识别的可疑示例分类为三类：分布外（OOD）示例，边界（BND）（BND）示例和较高分布错误分类（IDM）地区的示例。通过广泛的实验，我们阐明了现有的UQ方法，并表明了模型之间不确定性的原因有所不同。此外，我们展示了建议的框架如何利用分类的示例来提高预测性能。

Calibration is a popular framework to evaluate whether a classifier knows when it does not know - i.e., its predictive probabilities are a good indication of how likely a prediction is to be correct. Correctness is commonly estimated against the human majority class. Recently, calibration to human majority has been measured on tasks where humans inherently disagree about which class applies. We show that measuring calibration to human majority given inherent disagreements is theoretically problematic, demonstrate this empirically on the ChaosNLI dataset, and derive several instance-level measures of calibration that capture key statistical properties of human judgements - class frequency, ranking and entropy.

最近，深度学习中的不确定性估计已成为提高安全至关重要应用的可靠性和鲁棒性的关键领域。尽管有许多提出的方法要么关注距离感知模型的不确定性，要么是分布式检测的不确定性，要么是针对分布校准的输入依赖性标签不确定性，但这两种类型的不确定性通常都是必要的。在这项工作中，我们提出了用于共同建模模型和数据不确定性的HETSNGP方法。我们表明，我们提出的模型在这两种类型的不确定性之间提供了有利的组合，因此在包括CIFAR-100C，ImagEnet-C和Imagenet-A在内的一些具有挑战性的分发数据集上优于基线方法。此外，我们提出了HETSNGP Ensemble，这是我们方法的结合版本，该版本还对网络参数的不确定性进行建模，并优于其他集合基线。