Ensemble methods which average over multiple neural network predictions are a simple approach to improve a model's calibration and robustness. Similarly, data augmentation techniques, which encode prior information in the form of invariant feature transformations, are effective for improving calibration and robustness. In this paper, we show a surprising pathology: combining ensembles and data augmentation can harm model calibration. This leads to a trade-off in practice, whereby improved accuracy by combining the two techniques comes at the expense of calibration. On the other hand, selecting only one of the techniques ensures good uncertainty estimates at the expense of accuracy. We investigate this pathology and identify a compounding under-confidence among methods which marginalize over sets of weights and data augmentation techniques which soften labels. Finally, we propose a simple correction, achieving the best of both worlds with significant accuracy and calibration gains over using only ensembles or data augmentation individually. Applying the correction produces new state-of-the art in uncertainty calibration across CIFAR-10, CIFAR-100, and ImageNet.
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…
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…
Impact of Data Augmentation on Confidence Calibration in Melanoma Classification
Technological University Dublin（都柏林理工大学）；Research Ireland Centre for Research Training in Machine Learning（爱尔兰研究机器学习研究培训中心）；University College Dublin（都柏林大学学院）；Dublin City University（都柏林城市大学）
Diverse data augmentation strategies are a natural approach to improving robustness in computer vision models against unforeseen shifts in data distribution. However, the ability to tailor such strategies to inoculate a model against specific classes of corruptions or attacks -- without incurring substantial losses in robustness against other classes of corruptions -- remains elusive. In this work, we successfully harden a model against Fourier-based attacks, while producing superior-to-AugMix accuracy and calibration results on both the CIFAR-10-C and CIFAR-100-C datasets; classification erro…
A wide breadth of research has devised data augmentation approaches that can improve both accuracy and generalization performance for neural networks. However, augmented data can end up being far from the clean training data and what is the appropriate label is less clear. Despite this, most existing work simply uses one-hot labels for augmented data. In this paper, we show re-using one-hot labels for highly distorted data might run the risk of adding noise and degrading accuracy and calibration. To mitigate this, we propose a generic method AutoLabel to automatically learn the confidence in t…
University of Cambridge（剑桥大学）；UC San Diego（加州大学圣地亚哥分校）；NVIDIA（英伟达）；Apple（苹果公司）
Calibration is a fundamental property of a good predictive model: it requires that the model predicts correctly in proportion to its confidence. Modern neural networks, however, provide no strong guarantees on their calibration -- and can be either poorly calibrated or well-calibrated depending on the setting. It is currently unclear which factors contribute to good calibration (architecture, data augmentation, overparameterization, etc), though various claims exist in the literature. We propose a systematic way to study the calibration error: by decomposing it into (1) calibration error on th…
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 how…
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…
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…
Attention-Based Synthetic Data Generation for Calibration-Enhanced Survival Analysis: A Case Study for Chronic Kidney Disease Using Electronic Health Records
Centre for Big Data Research in Health (CBDRH)（健康大数据研究中心）；The University of New South Wales（新南威尔士大学）
Lawrence Livermore National Laboratory（劳伦斯利弗莫尔国家实验室）
While directly fine-tuning (FT) large-scale, pretrained models on task-specific data is well-known to induce strong in-distribution task performance, recent works have demonstrated that different adaptation protocols, such as linear probing (LP) prior to FT, can improve out-of-distribution generalization. However, the design space of such adaptation protocols remains under-explored and the evaluation of such protocols has primarily focused on distribution shifts. Therefore, in this work, we evaluate common adaptation protocols across distributions shifts and machine learning safety metrics (e.…
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…
Drowsiness reduces concentration and increases response time, which causes fatal road accidents. Monitoring drivers' drowsiness levels by electroencephalogram (EEG) and taking action may prevent road accidents. EEG signals effectively monitor the driver's mental state as they can monitor brain dynamics. However, calibration is required in advance because EEG signals vary between and within subjects. Because of the inconvenience, calibration has reduced the accessibility of the brain-computer interface (BCI). Developing a generalized classification model is similar to domain generalization, whi…
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）
Modern deep neural networks can produce badly calibrated predictions, especially when train and test distributions are mismatched. Training an ensemble of models and averaging their predictions can help alleviate these issues. We propose a simple technique to improve calibration, using a different data augmentation for each ensemble member. We additionally use the idea of `mixing' un-augmented and augmented inputs to improve calibration when test and training distributions are the same. These simple techniques improve calibration and accuracy over strong baselines on the CIFAR10 and CIFAR100 b…
Aalto University（阿尔托大学）；University of Manchester（曼彻斯特大学）
We introduce implicit Bayesian neural networks, a simple and scalable approach for uncertainty representation in deep learning. Standard Bayesian approach to deep learning requires the impractical inference of the posterior distribution over millions of parameters. Instead, we propose to induce a distribution that captures the uncertainty over neural networks by augmenting each layer's inputs with latent variables. We present appropriate input distributions and demonstrate state-of-the-art performance in terms of calibration, robustness and uncertainty characterisation over large-scale, multi-…
We propose a new scalable multi-class Gaussian process classification approach building on a novel modified softmax likelihood function. The new likelihood has two benefits: it leads to well-calibrated uncertainty estimates and allows for an efficient latent variable augmentation. The augmented model has the advantage that it is conditionally conjugate leading to a fast variational inference method via block coordinate ascent updates. Previous approaches suffered from a trade-off between uncertainty calibration and speed. Our experiments show that our method leads to well-calibrated uncertaint…
RMIT University（皇家墨尔本理工大学）；University College London（伦敦大学学院）
In this paper, we present a robust spherical harmonics approach for the classification of point cloud-based objects. Spherical harmonics have been used for classification over the years, with several frameworks existing in the literature. These approaches use variety of spherical harmonics based descriptors to classify objects. We first investigated these frameworks robustness against data augmentation, such as outliers and noise, as it has not been studied before. Then we propose a spherical convolution neural network framework for robust object classification. The proposed framework uses the…
VUNO, Inc.（VUNO公司）；Pukyong National University（釜庆国立大学）；Korea University（高丽大学）
While pre-trained language models (PLMs) have become a de-facto standard promoting the accuracy of text classification tasks, recent studies find that PLMs often predict over-confidently. Although various calibration methods have been proposed, such as ensemble learning and data augmentation, most of the methods have been verified in computer vision benchmarks rather than in PLM-based text classification tasks. In this paper, we present an empirical study on confidence calibration for PLMs, addressing three categories, including confidence penalty losses, data augmentations, and ensemble metho…
HarmAug: Effective Data Augmentation for Knowledge Distillation of Safety Guard Models
KAIST（韩国科学技术院）；Theori；Université de Montréal（蒙特利尔大学）；Mila – Québec AI Institute（米拉-魁北克人工智能研究所）；Technische Hochschule Nürnberg Georg Simon Ohm（纽伦堡应用技术大学）；CIFAR AI Chair（加拿大高级研究所人工智能主席项目）
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…
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…
Channel Reflection: Knowledge-Driven Data Augmentation for EEG-Based Brain-Computer Interfaces
Huazhong University of Science and Technology（华中科技大学）；Shenzhen Huazhong University of Science and Technology Research Institute（华中科技大学深圳研究院）；China Electronic System Technology Co., Ltd.（中国电子系统技术有限公司）；School of Artificial Intelligence and Automation, Huazhong University of Science and Technology（华中科技大学人工智能与自动化学院）；School of Civil and Hydraulic Engineering, Huazhong University of Science and Technology（华中科技大学土木与水利工程学院）
Recent works have shown that self-supervised learning can achieve remarkable robustness when integrated with adversarial training (AT). However, the robustness gap between supervised AT (sup-AT) and self-supervised AT (self-AT) remains significant. Motivated by this observation, we revisit existing self-AT methods and discover an inherent dilemma that affects self-AT robustness: either strong or weak data augmentations are harmful to self-AT, and a medium strength is insufficient to bridge the gap. To resolve this dilemma, we propose a simple remedy named DYNACL (Dynamic Adversarial Contrastiv…
In computer vision, it is standard practice to draw a single sample from the data augmentation procedure for each unique image in the mini-batch. However recent work has suggested drawing multiple samples can achieve higher test accuracies. In this work, we provide a detailed empirical evaluation of how the number of augmentation samples per unique image influences model performance on held out data when training deep ResNets. We demonstrate drawing multiple samples per image consistently enhances the test accuracy achieved for both small and large batch training. Crucially, this benefit arise…
Beijing University of Posts and Telecommunications（北京邮电大学）
Contrastive learning (CL), which can extract the information shared between different contrastive views, has become a popular paradigm for vision representation learning. Inspired by the success in computer vision, recent work introduces CL into graph modeling, dubbed as graph contrastive learning (GCL). However, generating contrastive views in graphs is more challenging than that in images, since we have little prior knowledge on how to significantly augment a graph without changing its labels. We argue that typical data augmentation techniques (e.g., edge dropping) in GCL cannot generate div…
General Motors - Advanced Technical Center Israel（通用汽车以色列先进技术中心）；Tel-Aviv University（特拉维夫大学）
While camera and LiDAR processing have been revolutionized since the introduction of deep learning, radar processing still relies on classical tools. In this paper, we introduce a deep learning approach for radar processing, working directly with the radar complex data. To overcome the lack of radar labeled data, we rely in training only on the radar calibration data and introduce new radar augmentation techniques. We evaluate our method on the radar 4D detection task and demonstrate superior performance compared to the classical approaches while keeping real-time performance. Applying deep le…
Korea Advanced Institute of Science and Technology (KAIST)（韩国科学技术院）；University of Michigan（密歇根大学）；University of California, Berkeley（加州大学伯克利分校）；Amazon Web Services（亚马逊云科技）；LG AI Research（LG人工智能研究院）
Recent unsupervised representation learning methods have shown to be effective in a range of vision tasks by learning representations invariant to data augmentations such as random cropping and color jittering. However, such invariance could be harmful to downstream tasks if they rely on the characteristics of the data augmentations, e.g., location- or color-sensitive. This is not an issue just for unsupervised learning; we found that this occurs even in supervised learning because it also learns to predict the same label for all augmented samples of an instance. To avoid such failures and obt…
Shien-Ming Wu School of Intelligent Engineering, South China University of Technology（华南理工大学吴贤铭智能工程学院）
The cross-subject application of EEG-based brain-computer interface (BCI) has always been limited by large individual difference and complex characteristics that are difficult to perceive. Therefore, it takes a long time to collect the training data of each user for calibration. Even transfer learning method pre-training with amounts of subject-independent data cannot decode different EEG signal categories without enough subject-specific data. Hence, we proposed a cross-subject EEG classification framework with a generative adversarial networks (GANs) based method named common spatial GAN (CS-…
