On Calibration of Mixup Training for Deep Neural Networks
PRHLT Research Center, Universitat Politècnica de València（PRHLT研究中心，瓦伦西亚理工大学）；AUDIAS - Audio, Data Intelligence and Speech, Universidad Autónoma de Madrid（AUDIAS - 音频、数据智能与语音，马德里自治大学）
Deep Neural Networks (DNN) represent the state of the art in many tasks. However, due to their overparameterization, their generalization capabilities are in doubt and still a field under study. Consequently, DNN can overfit and assign overconfident predictions -- effects that have been shown to affect the calibration of the confidences assigned to unseen data. Data Augmentation (DA) strategies have been proposed to regularize these models, being Mixup one of the most popular due to its ability to improve the accuracy, the uncertainty quantification and the calibration of DNN. In this work however we argue and provide empirical evidence that, due to its fundamentals, Mixup does not necessarily improve calibration. Based on our observations we propose a new loss function that improves the calibration, and also sometimes the accuracy, of DNN trained with this DA technique. Our loss is inspired by Bayes decision theory and introduces a new training framework for designing losses for probabilistic modelling. We provide state-of-the-art accuracy with consistent improvements in calibration performance. Appendix and code are provided here: https://github.com/jmaronas/calibration_MixupDNN_ARCLoss.pytorch.git
Los Alamos National Laboratory（洛斯阿拉莫斯国家实验室）；University of Washington（华盛顿大学）
Mixup~\cite{zhang2017mixup} is a recently proposed method for training deep neural networks where additional samples are generated during training by convexly combining random pairs of images and their associated labels. While simple to implement, it has been shown to be a surprisingly effective method of data augmentation for image classification: DNNs trained with mixup show noticeable gains in classification performance on a number of image classification benchmarks. In this work, we discuss a hitherto untouched aspect of mixup training -- the calibration and predictive uncertainty of model…
A well-calibrated neural model produces confidence (probability outputs) closely approximated by the expected accuracy. While prior studies have shown that mixup training as a data augmentation technique can improve model calibration on image classification tasks, little is known about using mixup for model calibration on natural language understanding (NLU) tasks. In this paper, we explore mixup for model calibration on several NLU tasks and propose a novel mixup strategy for pre-trained language models that improves model calibration further. Our proposed mixup is guided by both the Area Und…
Improving the generalization of deep networks is an important open challenge, particularly in domains without plentiful data. The mixup algorithm improves generalization by linearly interpolating a pair of examples and their corresponding labels. These interpolated examples augment the original training set. Mixup has shown promising results in various classification tasks, but systematic analysis of mixup in regression remains underexplored. Using mixup directly on regression labels can result in arbitrarily incorrect labels. In this paper, we propose a simple yet powerful algorithm, C-Mixup,…
Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院）；AITRICS
Graph-structured datasets usually have irregular graph sizes and connectivities, rendering the use of recent data augmentation techniques, such as Mixup, difficult. To tackle this challenge, we present the first Mixup-like graph augmentation method at the graph-level called Graph Transplant, which mixes irregular graphs in data space. To be well defined on various scales of the graph, our method identifies the sub-structure as a mix unit that can preserve the local information. Since the mixup-based methods without special consideration of the context are prone to generate noisy samples, our m…
It has been widely recognized that adversarial examples can be easily crafted to fool deep networks, which mainly root from the locally non-linear behavior nearby input examples. Applying mixup in training provides an effective mechanism to improve generalization performance and model robustness against adversarial perturbations, which introduces the globally linear behavior in-between training examples. However, in previous work, the mixup-trained models only passively defend adversarial attacks in inference by directly classifying the inputs, where the induced global linearity is not well ex…
The success of deep learning relies on the availability of large-scale annotated data sets, the acquisition of which can be costly, requiring expert domain knowledge. Semi-supervised learning (SSL) mitigates this challenge by exploiting the behavior of the neural function on large unlabeled data. The smoothness of the neural function is a commonly used assumption exploited in SSL. A successful example is the adoption of mixup strategy in SSL that enforces the global smoothness of the neural function by encouraging it to behave linearly when interpolating between training examples. Despite its …
Arizona State University（亚利桑那州立大学）；Lawrence Livermore National Labs（劳伦斯利弗莫尔国家实验室）；IBM Research AI（IBM研究院人工智能部门）
The hypothesis that sub-network initializations (lottery) exist within the initializations of over-parameterized networks, which when trained in isolation produce highly generalizable models, has led to crucial insights into network initialization and has enabled efficient inferencing. Supervised models with uncalibrated confidences tend to be overconfident even when making wrong prediction. In this paper, for the first time, we study how explicit confidence calibration in the over-parameterized network impacts the quality of the resulting lottery tickets. More specifically, we incorporate a s…
Aalto Univeristy（阿尔托大学）；Montréal Institute for Learning Algorithms（蒙特利尔学习算法研究所）；Sharif University of Technology（谢里夫理工大学）；Facebook AI Research（Facebook人工智能研究院）
Deep neural networks excel at learning the training data, but often provide incorrect and confident predictions when evaluated on slightly different test examples. This includes distribution shifts, outliers, and adversarial examples. To address these issues, we propose Manifold Mixup, a simple regularizer that encourages neural networks to predict less confidently on interpolations of hidden representations. Manifold Mixup leverages semantic interpolations as additional training signal, obtaining neural networks with smoother decision boundaries at multiple levels of representation. As a resu…
Deep Learning methods are known to suffer from calibration issues: they typically produce over-confident estimates. These problems are exacerbated in the low data regime. Although the calibration of probabilistic models is well studied, calibrating extremely over-parametrized models in the low-data regime presents unique challenges. We show that deep-ensembles do not necessarily lead to improved calibration properties. In fact, we show that standard ensembling methods, when used in conjunction with modern techniques such as mixup regularization, can lead to less calibrated models. This text ex…
Bosch Center for Artificial Intelligence（博世人工智能中心）；University of Stuttgart（斯图加特大学）
Uncertainty estimates help to identify ambiguous, novel, or anomalous inputs, but the reliable quantification of uncertainty has proven to be challenging for modern deep networks. In order to improve uncertainty estimation, we propose On-Manifold Adversarial Data Augmentation or OMADA, which specifically attempts to generate the most challenging examples by following an on-manifold adversarial attack path in the latent space of an autoencoder-based generative model that closely approximates decision boundaries between two or more classes. On a variety of datasets as well as on multiple diverse…
Rutgers University（罗格斯大学）；Harvard University（哈佛大学）；National University of Singapore（新加坡国立大学）；Stanford University（斯坦福大学）
In many machine learning applications, it is important for the model to provide confidence scores that accurately capture its prediction uncertainty. Although modern learning methods have achieved great success in predictive accuracy, generating calibrated confidence scores remains a major challenge. Mixup, a popular yet simple data augmentation technique based on taking convex combinations of pairs of training examples, has been empirically found to significantly improve confidence calibration across diverse applications. However, when and how Mixup helps calibration is still a mystery. In th…
In this work, we propose data augmentation methods for embeddings from pre-trained deep learning models that take a weighted combination of a pair of input embeddings, as inspired by Mixup, and combine such augmentation with extra label softening. These methods are shown to significantly increase classification accuracy, reduce training time, and improve confidence calibration of a downstream model that is trained with them. As a result of such improved confidence calibration, the model output can be more intuitively interpreted and used to accurately identify out-of-distribution data by apply…
MixUp is a computer vision data augmentation technique that uses convex interpolations of input data and their labels to enhance model generalization during training. However, the application of MixUp to the natural language understanding (NLU) domain has been limited, due to the difficulty of interpolating text directly in the input space. In this study, we propose MixUp methods at the Input, Manifold, and sentence embedding levels for the transformer architecture, and apply them to finetune the BERT model for a diverse set of NLU tasks. We find that MixUp can improve model performance, as we…
Improving Mixup Calibration with Wasserstein Distributionally Robust Optimization
Department of Mathematics & Statistics（数学与统计学系）；Boston University（波士顿大学）；Department of Electrical and Computer Engineering, Division of Systems Engineering（电气与计算机工程系，系统工程 division）；Department of Biomedical Engineering（生物医学工程系）；Faculty of Computing & Data Sciences（计算与数据科学学院）
Deep neural networks are susceptible to adversarially crafted, small and imperceptible changes in the natural inputs. The most effective defense mechanism against these examples is adversarial training which constructs adversarial examples during training by iterative maximization of loss. The model is then trained to minimize the loss on these constructed examples. This min-max optimization requires more data, larger capacity models, and additional computing resources. It also degrades the standard generalization performance of a model. Can we achieve robustness more efficiently? In this work…
Lawrence Livermore National Laboratory（劳伦斯利弗莫尔国家实验室）
The role of uncertainty quantification (UQ) in deep learning has become crucial with growing use of predictive models in high-risk applications. Though a large class of methods exists for measuring deep uncertainties, in practice, the resulting estimates are found to be poorly calibrated, thus making it challenging to translate them into actionable insights. A common workaround is to utilize a separate recalibration step, which adjusts the estimates to compensate for the miscalibration. Instead, we propose to repurpose the heteroscedastic regression objective as a surrogate for calibration and…
Mixed Sample Data Augmentation (MSDA) has received increasing attention in recent years, with many successful variants such as MixUp and CutMix. By studying the mutual information between the function learned by a VAE on the original data and on the augmented data we show that MixUp distorts learned functions in a way that CutMix does not. We further demonstrate this by showing that MixUp acts as a form of adversarial training, increasing robustness to attacks such as Deep Fool and Uniform Noise which produce examples similar to those generated by MixUp. We argue that this distortion prevents …
Deep neural networks have been shown to be highly miscalibrated. often they tend to be overconfident in their predictions. It poses a significant challenge for safety-critical systems to utilise deep neural networks (DNNs), reliably. Many recently proposed approaches to mitigate this have demonstrated substantial progress in improving DNN calibration. However, they hardly touch upon refinement, which historically has been an essential aspect of calibration. Refinement indicates separability of a network's correct and incorrect predictions. This paper presents a theoretically and empirically su…
Lawrence Livermore National Laboratories（劳伦斯利弗莫尔国家实验室）
This paper studies the problem of post-hoc calibration of machine learning classifiers. We introduce the following desiderata for uncertainty calibration: (a) accuracy-preserving, (b) data-efficient, and (c) high expressive power. We show that none of the existing methods satisfy all three requirements, and demonstrate how Mix-n-Match calibration strategies (i.e., ensemble and composition) can help achieve remarkably better data-efficiency and expressive power while provably maintaining the classification accuracy of the original classifier. Mix-n-Match strategies are generic in the sense that…
Deep neural networks have found widespread adoption in solving complex tasks ranging from image recognition to natural language processing. However, these networks make confident mispredictions when presented with data that does not belong to the training distribution, i.e. out-of-distribution (OoD) samples. In this paper we explore whether the property of Vicinal Risk Minimization (VRM) to smoothly interpolate between different class boundaries helps to train better OoD detectors. We apply VRM to existing OoD detection techniques and show their improved performance. We observe that existing O…
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Deep neural network (DNN) classifiers are often overconfident, producing miscalibrated class probabilities. In high-risk applications like healthcare, practitioners require $\textit{fully calibrated}$ probability predictions for decision-making. That is, conditioned on the prediction $\textit{vector}$, $\textit{every}$ class' probability should be close to the predicted value. Most existing calibration methods either lack theoretical guarantees for producing calibrated outputs, reduce classification accuracy in the process, or only calibrate the predicted class. This paper proposes a new Kerne…
Beyond Overconfidence: Foundation Models Redefine Calibration in Deep Neural Networks
Goethe University Frankfurt（弗赖堡歌德大学）；German Cancer Consortium (DKTK)（德国癌症联盟（DKTK））；German Cancer Research Center (DKFZ)（德国癌症研究中心（DKFZ））；Frankfurt Cancer Institute（法兰克福癌症研究所）
Seoul National University（首尔国立大学）；Neural Processing Research Center（神经处理研究中心）
While deep neural networks show great performance on fitting to the training distribution, improving the networks' generalization performance to the test distribution and robustness to the sensitivity to input perturbations still remain as a challenge. Although a number of mixup based augmentation strategies have been proposed to partially address them, it remains unclear as to how to best utilize the supervisory signal within each input data for mixup from the optimization perspective. We propose a new perspective on batch mixup and formulate the optimal construction of a batch of mixup data …
MaLGa - University of Genova, Italy（马尔加-热那亚大学，意大利）；Google Research - Brain team, Accra（谷歌研究-大脑团队，阿克拉）；Google Research - Brain team, Berlin（谷歌研究-大脑团队，柏林）；Google Research - Brain team, Paris（谷歌研究-大脑团队，巴黎）
Large deep neural networks are powerful, but exhibit undesirable behaviors such as memorization and sensitivity to adversarial examples. In this work, we propose mixup, a simple learning principle to alleviate these issues. In essence, mixup trains a neural network on convex combinations of pairs of examples and their labels. By doing so, mixup regularizes the neural network to favor simple linear behavior in-between training examples. Our experiments on the ImageNet-2012, CIFAR-10, CIFAR-100, Google commands and UCI datasets show that mixup improves the generalization of state-of-the-art neur…
Recently, there has been a growing interest in applying machine learning methods to problems in engineering mechanics. In particular, there has been significant interest in applying deep learning techniques to predicting the mechanical behavior of heterogeneous materials and structures. Researchers have shown that deep learning methods are able to effectively predict mechanical behavior with low error for systems ranging from engineered composites, to geometrically complex metamaterials, to heterogeneous biological tissue. However, there has been comparatively little attention paid to deep lea…
Deep Neural Network Calibration by Reducing Classifier Shift with Stochastic Masking
School of Artificial Intelligence, Jilin University（吉林大学人工智能学院）；CSIRO's Data61 and Monash University (adjunct)（CSIRO的数据61与墨尔本大学（兼职））；King Abdullah University of Science and Technology（国王阿卜杜勒阿齐兹科技大学）
Deep Learning-Enhanced Calibration of the Heston Model: A Unified Framework
Department of Applied Mathematics, Faculty of Mathematical Sciences, University of Guilan, P. O. Box: 41938-1914, Rasht, Iran（应用数学系，数学科学学院，加兹万大学，伊朗拉什特邮编41938-1914）；Department of Mathematics, Faculty of Mathematical Sciences, Alzahra University, Tehran, Iran（数学系，数学科学学院，阿尔扎哈拉大学，伊朗德黑兰）；Department of Economics and Business, Universidad de Almeria, Spain（经济与商业系，阿尔梅里亚大学，西班牙）
Beyond Dark Knowledge: Mixup-Based Distillation for Reliable Predictions
ITS Lab, Institute of Computer Science, University of Tartu（塔尔图大学计算机科学学院ITS实验室）；LITIS, Université de Rouen（鲁昂大学LITIS实验室）；LITIS, INSA de Rouen（鲁昂国立应用科学学院LITIS实验室）
University of Florida（佛罗里达大学）；United States Air Force Research Laboratory（美国空军研究实验室）
Deep learning has achieved the state-of-the-art performance across medical imaging tasks; however, model calibration is often not considered. Uncalibrated models are potentially dangerous in high-risk applications since the user does not know when they will fail. Therefore, this paper proposes a novel domain-aware loss function to calibrate deep learning models. The proposed loss function applies a class-wise penalty based on the similarity between classes within a given target domain. Thus, the approach improves the calibration while also ensuring that the model makes less risky errors even w…
National Research Council Canada（加拿大国家研究委员会）；University of Ottawa（渥太华大学）；Beihang University（北京航空航天大学）
MixUp is a recently proposed data-augmentation scheme, which linearly interpolates a random pair of training examples and correspondingly the one-hot representations of their labels. Training deep neural networks with such additional data is shown capable of significantly improving the predictive accuracy of the current art. The power of MixUp, however, is primarily established empirically and its working and effectiveness have not been explained in any depth. In this paper, we develop an understanding for MixUp as a form of "out-of-manifold regularization", which imposes certain "local linear…
Deep learning architectures have achieved promising results in different areas (e.g., medicine, agriculture, and security). However, using those powerful techniques in many real applications becomes challenging due to the large labeled collections required during training. Several works have pursued solutions to overcome it by proposing strategies that can learn more for less, e.g., weakly and semi-supervised learning approaches. As these approaches do not usually address memorization and sensitivity to adversarial examples, this paper presents three deep metric learning approaches combined wi…
Transferring knowledge from large source datasets is an effective way to fine-tune the deep neural networks of the target task with a small sample size. A great number of algorithms have been proposed to facilitate deep transfer learning, and these techniques could be generally categorized into two groups - Regularized Learning of the target task using models that have been pre-trained from source datasets, and Multitask Learning with both source and target datasets to train a shared backbone neural network. In this work, we aim to improve the multitask paradigm for deep transfer learning via …
Korea University（高丽大学）；Meta Reality Labs（Meta现实实验室）
Data augmentation is key to improving the generalization ability of deep learning models. Mixup is a simple and widely-used data augmentation technique that has proven effective in alleviating the problems of overfitting and data scarcity. Also, recent studies of saliency-aware Mixup in the image domain show that preserving discriminative parts is beneficial to improving the generalization performance. However, these Mixup-based data augmentations are underexplored in 3D vision, especially in point clouds. In this paper, we propose SageMix, a saliency-guided Mixup for point clouds to preserve …
The Mixup method (Zhang et al. 2018), which uses linearly interpolated data, has emerged as an effective data augmentation tool to improve generalization performance and the robustness to adversarial examples. The motivation is to curtail undesirable oscillations by its implicit model constraint to behave linearly at in-between observed data points and promote smoothness. In this work, we formally investigate this premise, propose a way to explicitly impose smoothness constraints, and extend it to incorporate with implicit model constraints. First, we derive a new function class composed of ke…
This paper proposes a novel formulation of prototypical loss with mixup for speaker verification. Mixup is a simple yet efficient data augmentation technique that fabricates a weighted combination of random data point and label pairs for deep neural network training. Mixup has attracted increasing attention due to its ability to improve robustness and generalization of deep neural networks. Although mixup has shown success in diverse domains, most applications have centered around closed-set classification tasks. In this work, we propose contrastive-mixup, a novel augmentation strategy that le…
University of Ottawa（渥太华大学）；National Research Council Canada（加拿大国家研究委员会）
Mixup, which creates synthetic training instances by linearly interpolating random sample pairs, is a simple and yet effective regularization technique to boost the performance of deep models trained with SGD. In this work, we report a previously unobserved phenomenon in Mixup training: on a number of standard datasets, the performance of Mixup-trained models starts to decay after training for a large number of epochs, giving rise to a U-shaped generalization curve. This behavior is further aggravated when the size of original dataset is reduced. To help understand such a behavior of Mixup, we…
The University of Hong Kong（香港大学）；Carnegie Mellon University（卡内基梅隆大学）；University of California, Los Angeles（加州大学洛杉矶分校）
Mixup, a simple data augmentation method that randomly mixes two data points via linear interpolation, has been extensively applied in various deep learning applications to gain better generalization. However, the theoretical underpinnings of its efficacy are not yet fully understood. In this paper, we aim to seek a fundamental understanding of the benefits of Mixup. We first show that Mixup using different linear interpolation parameters for features and labels can still achieve similar performance to the standard Mixup. This indicates that the intuitive linearity explanation in Zhang et al.,…
MixUp is an effective data augmentation method to regularize deep neural networks via random linear interpolations between pairs of samples and their labels. It plays an important role in model regularization, semi-supervised learning and domain adaption. However, despite its empirical success, its deficiency of randomly mixing samples has poorly been studied. Since deep networks are capable of memorizing the entire dataset, the corrupted samples generated by vanilla MixUp with a badly chosen interpolation policy will degrade the performance of networks. To overcome the underfitting by corrupt…
Simon Fraser University（西蒙弗雷泽大学）；Hinge Health（Hinge Health公司）
Modern deep learning training procedures rely on model regularization techniques such as data augmentation methods, which generate training samples that increase the diversity of data and richness of label information. A popular recent method, mixup, uses convex combinations of pairs of original samples to generate new samples. However, as we show in our experiments, mixup can produce undesirable synthetic samples, where the data is sampled off the manifold and can contain incorrect labels. We propose $ζ$-mixup, a generalization of mixup with provably and demonstrably desirable properties that…
Seoul National University（首尔国立大学）；Neural Processing Research Center（神经处理研究中心）
While deep neural networks achieve great performance on fitting the training distribution, the learned networks are prone to overfitting and are susceptible to adversarial attacks. In this regard, a number of mixup based augmentation methods have been recently proposed. However, these approaches mainly focus on creating previously unseen virtual examples and can sometimes provide misleading supervisory signal to the network. To this end, we propose Puzzle Mix, a mixup method for explicitly utilizing the saliency information and the underlying statistics of the natural examples. This leads to a…
