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 linearity" constraints on the model's input space beyond the data manifold. This analysis enables us to identify a limitation of MixUp, which we call "manifold intrusion". In a nutshell, manifold intrusion in MixUp is a form of under-fitting resulting from conflicts between the synthetic labels of the mixed-up examples and the labels of original training data. Such a phenomenon usually happens when the parameters controlling the generation of mixing policies are not sufficiently fine-tuned on the training data. To address this issue, we propose a novel adaptive version of MixUp, where the mixing policies are automatically learned from the data using an additional network and objective function designed to avoid manifold intrusion. The proposed regularizer, AdaMixUp, is empirically evaluated on several benchmark datasets. Extensive experiments demonstrate that AdaMixUp improves upon MixUp when applied to the current art of deep classification models.
Pennsylvania State University（宾夕法尼亚州立大学）；Sant’Anna School of Advanced Studies（圣安娜高等研究院）
In this paper we propose $ε$-Consistent Mixup ($ε$mu). $ε$mu is a data-based structural regularization technique that combines Mixup's linear interpolation with consistency regularization in the Mixup direction, by compelling a simple adaptive tradeoff between the two. This learnable combination of consistency and interpolation induces a more flexible structure on the evolution of the response across the feature space and is shown to improve semi-supervised classification accuracy on the SVHN and CIFAR10 benchmark datasets, yielding the largest gains in the most challenging low label-availabil…
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）
Machine learning techniques are used in a wide range of domains. However, machine learning models often suffer from the problem of over-fitting. Many data augmentation methods have been proposed to tackle such a problem, and one of them is called mixup. Mixup is a recently proposed regularization procedure, which linearly interpolates a random pair of training examples. This regularization method works very well experimentally, but its theoretical guarantee is not adequately discussed. In this study, we aim to discover why mixup works well from the aspect of the statistical learning theory.
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 …
Mixup - a neural network regularization technique based on linear interpolation of labeled sample pairs - has stood out by its capacity to improve model's robustness and generalizability through a surprisingly simple formalism. However, its extension to the field of object detection remains unclear as the interpolation of bounding boxes cannot be naively defined. In this paper, we propose to leverage the inherent region mapping structure of anchors to introduce a mixup-driven training regularization for region proposal based object detectors. The proposed method is benchmarked on standard data…
While novel computer vision architectures are gaining traction, the impact of model architectures is often related to changes or exploring in training methods. Identity mapping-based architectures ResNets and DenseNets have promised path-breaking results in the image classification task and are go-to methods for even now if the data given is fairly limited. Considering the ease of training with limited resources this work revisits the ResNets and improves the ResNet50 \cite{resnets} by using mixup data-augmentation as regularization and tuning the hyper-parameters.
Mila - Quebec AI Institute（米拉-魁北克人工智能研究所）；McGill University（麦吉尔大学）；Université de Montréal（蒙特利尔大学）
It is common in computer vision to be confronted with domain shift: images which have the same class but different acquisition conditions. In domain adaptation (DA), one wants to classify unlabeled target images using source labeled images. Unfortunately, deep neural networks trained on a source training set perform poorly on target images which do not belong to the training domain. One strategy to improve these performances is to align the source and target image distributions in an embedded space using optimal transport (OT). However OT can cause negative transfer, i.e. aligning samples with…
University of California, Merced（加州大学默塞德分校）；eBay Inc.（eBay公司）；NEC Labs America（美国NEC实验室）；Google Research（谷歌研究院）
Obtaining object response maps is one important step to achieve weakly-supervised semantic segmentation using image-level labels. However, existing methods rely on the classification task, which could result in a response map only attending on discriminative object regions as the network does not need to see the entire object for optimizing the classification loss. To tackle this issue, we propose a principled and end-to-end train-able framework to allow the network to pay attention to other parts of the object, while producing a more complete and uniform response map. Specifically, we introdu…
Mixup is a recent regularizer for current deep classification networks. Through training a neural network on convex combinations of pairs of examples and their labels, it imposes locally linear constraints on the model's input space. However, such strict linear constraints often lead to under-fitting which degrades the effects of regularization. Noticeably, this issue is getting more serious when the resource is extremely limited. To address these issues, we propose the Adversarial Mixing Policy (AMP), organized in a min-max-rand formulation, to relax the Locally Linear Constraints in Mixup. S…
Mixup is a popular data augmentation technique based on taking convex combinations of pairs of examples and their labels. This simple technique has been shown to substantially improve both the robustness and the generalization of the trained model. However, it is not well-understood why such improvement occurs. In this paper, we provide theoretical analysis to demonstrate how using Mixup in training helps model robustness and generalization. For robustness, we show that minimizing the Mixup loss corresponds to approximately minimizing an upper bound of the adversarial loss. This explains why m…
Few-shot learning algorithms aim to learn model parameters capable of adapting to unseen classes with the help of only a few labeled examples. A recent regularization technique - Manifold Mixup focuses on learning a general-purpose representation, robust to small changes in the data distribution. Since the goal of few-shot learning is closely linked to robust representation learning, we study Manifold Mixup in this problem setting. Self-supervised learning is another technique that learns semantically meaningful features, using only the inherent structure of the data. This work investigates th…
Mixup is a data-dependent regularization technique that consists in linearly interpolating input samples and associated outputs. It has been shown to improve accuracy when used to train on standard machine learning datasets. However, authors have pointed out that Mixup can produce out-of-distribution virtual samples and even contradictions in the augmented training set, potentially resulting in adversarial effects. In this paper, we introduce Local Mixup in which distant input samples are weighted down when computing the loss. In constrained settings we demonstrate that Local Mixup can create …
Zhejiang University（浙江大学）；University of Notre Dame（圣母大学）
In clinical practice, medical image interpretation often involves multi-labeled classification, since the affected parts of a patient tend to present multiple symptoms or comorbidities. Recently, deep learning based frameworks have attained expert-level performance on medical image interpretation, which can be attributed partially to large amounts of accurate annotations. However, manually annotating massive amounts of medical images is impractical, while automatic annotation is fast but imprecise (possibly introducing corrupted labels). In this work, we propose a new regularization approach, …
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…
Nordita, KTH Royal Institute of Technology（北欧理论物理研究所，KTH皇家理工学院）；Stockholm University（斯德哥尔摩大学）；University of Pittsburgh（匹兹堡大学）；ICSI（国际计算机科学研究所）；University of Toronto（多伦多大学）；UC Berkeley（加州大学伯克利分校）
Poznan University of Technology（波兹南理工大学）；Institute of Computing Science（计算机科学研究所）
Overfitting & underfitting and stable training are an important challenges in machine learning. Current approaches for these issues are mixup, SamplePairing and BC learning. In our work, we state the hypothesis that mixing many images together can be more effective than just two. Batchboost pipeline has three stages: (a) pairing: method of selecting two samples. (b) mixing: how to create a new one from two samples. (c) feeding: combining mixed samples with new ones from dataset into batch (with ratio $γ$). Note that sample that appears in our batch propagates with subsequent iterations with 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…
We propose the first unified theoretical analysis of mixed sample data augmentation (MSDA), such as Mixup and CutMix. Our theoretical results show that regardless of the choice of the mixing strategy, MSDA behaves as a pixel-level regularization of the underlying training loss and a regularization of the first layer parameters. Similarly, our theoretical results support that the MSDA training strategy can improve adversarial robustness and generalization compared to the vanilla training strategy. Using the theoretical results, we provide a high-level understanding of how different design choic…
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…
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…
Mixup is a regularization technique that artificially produces new samples using convex combinations of original training points. This simple technique has shown strong empirical performance, and has been heavily used as part of semi-supervised learning techniques such as mixmatch~\citep{berthelot2019mixmatch} and interpolation consistent training (ICT)~\citep{verma2019interpolation}. In this paper, we look at Mixup through a \emph{representation learning} lens in a semi-supervised learning setup. In particular, we study the role of Mixup in promoting linearity in the learned network represent…
Detecting the positions of human hands and objects-in-contact (hand-object detection) in each video frame is vital for understanding human activities from videos. For training an object detector, a method called Mixup, which overlays two training images to mitigate data bias, has been empirically shown to be effective for data augmentation. However, in hand-object detection, mixing two hand-manipulation images produces unintended biases, e.g., the concentration of hands and objects in a specific region degrades the ability of the hand-object detector to identify object boundaries. We propose a…
Carleton University（卡尔顿大学）；University of Ottawa（渥太华大学）
Using the shared-private paradigm and adversarial training has significantly improved the performances of multi-domain text classification (MDTC) models. However, there are two issues for the existing methods. First, instances from the multiple domains are not sufficient for domain-invariant feature extraction. Second, aligning on the marginal distributions may lead to fatal mismatching. In this paper, we propose a mixup regularized adversarial network (MRAN) to address these two issues. More specifically, the domain and category mixup regularizations are introduced to enrich the intrinsic fea…
NORCE Norwegian Research Centre AS（NORCE挪威研究中心）；University of Stavanger（斯塔万格大学）
Generalizing the application of machine learning models to situations where the statistical distribution of training and test data are different has been a complex problem. Our contributions in this paper are threefold: (1) we introduce an anchored-based Out of Distribution (OOD) Regression Mixup algorithm, leveraging manifold hidden state mixup and observation similarities to form a novel regularization penalty, (2) we provide a first of its kind, high resolution Distributed Acoustic Sensor (DAS) dataset that is suitable for testing OOD regression modelling, allowing other researchers to benc…
Graduate School of System Informatics, Kobe University（神户大学系统信息学研究生院）
Deep convolutional neural networks (CNNs) have achieved remarkable results in image processing tasks. However, their high expression ability risks overfitting. Consequently, data augmentation techniques have been proposed to prevent overfitting while enriching datasets. Recent CNN architectures with more parameters are rendering traditional data augmentation techniques insufficient. In this study, we propose a new data augmentation technique called random image cropping and patching (RICAP) which randomly crops four images and patches them to create a new training image. Moreover, RICAP mixes …
Deep learning approaches have shown great success in image classification tasks and can aid greatly towards the fast and reliable classification of pollen grain aerial imagery. However, often-times deep learning methods in the setting of natural images can suffer generalization problems and yield poor performance on unseen test distribution. In this work, we present and a robust deep learning framework that can generalize well for pollen grain aerobiological imagery classification. We develop a convolutional neural network-based pollen grain classification approach and combine some of the best…
Deep-learning based speech separation models confront poor generalization problem that even the state-of-the-art models could abruptly fail when evaluating them in mismatch conditions. To address this problem, we propose an easy-to-implement yet effective consistency based semi-supervised learning (SSL) approach, namely Mixup-Breakdown training (MBT). It learns a teacher model to "breakdown" unlabeled inputs, and the estimated separations are interpolated to produce more useful pseudo "mixup" input-output pairs, on which the consistency regularization could apply for learning a student model. …
Annot-Mix: Learning with Noisy Class Labels from Multiple Annotators via a Mixup Extension
University of Kassel（卡塞尔大学）；European Conference on Artificial Intelligence（欧洲人工智能会议）；Conference on Prestigious Applications of Intelligent Systems（智能系统 prestigious 应用会议）
New York University（纽约大学）；Facebook AI Research（Facebook人工智能研究）；CIFAR（加拿大高级研究所（CIFAR））
We propose a retrieval-augmented convolutional network and propose to train it with local mixup, a novel variant of the recently proposed mixup algorithm. The proposed hybrid architecture combining a convolutional network and an off-the-shelf retrieval engine was designed to mitigate the adverse effect of off-manifold adversarial examples, while the proposed local mixup addresses on-manifold ones by explicitly encouraging the classifier to locally behave linearly on the data manifold. Our evaluation of the proposed approach against five readily-available adversarial attacks on three datasets--…
Towards Optimization and Model Selection for Domain Generalization: A Mixup-guided Solution
Tsinghua University（清华大学）；Microsoft Research Asia（微软亚洲研究院）；National Engineering Research Center for Software Engineering, Peking University（北京大学软件工程国家工程研究中心）
National Research Council Canada（加拿大国家研究委员会）；University of Ottawa（渥太华大学）
We present a simple and yet effective interpolation-based regularization technique, aiming to improve the generalization of Graph Neural Networks (GNNs) on supervised graph classification. We leverage Mixup, an effective regularizer for vision, where random sample pairs and their labels are interpolated to create synthetic images for training. Unlike images with grid-like coordinates, graphs have arbitrary structure and topology, which can be very sensitive to any modification that alters the graph's semantic meanings. This posts two unanswered questions for Mixup-like regularization schemes: …
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…
Hong Kong Baptist University（香港浸会大学）；Nanyang Technological University（南洋理工大学）
Test-time training provides a new approach solving the problem of domain shift. In its framework, a test-time training phase is inserted between training phase and test phase. During test-time training phase, usually parts of the model are updated with test sample(s). Then the updated model will be used in the test phase. However, utilizing test samples for test-time training has some limitations. Firstly, it will lead to overfitting to the test-time procedure thus hurt the performance on the main task. Besides, updating part of the model without changing other parts will induce a mismatch pro…
University of Oxford（牛津大学）；Johns Hopkins University（约翰斯·霍普金斯大学）；The University of Hong Kong（香港大学）；ByteDance Inc.（字节跳动公司）
Modern deep networks can be better generalized when trained with noisy samples and regularization techniques. Mixup and CutMix have been proven to be effective for data augmentation to help avoid overfitting. Previous Mixup-based methods linearly combine images and labels to generate additional training data. However, this is problematic if the object does not occupy the whole image as we demonstrate in Figure 1. Correctly assigning the label weights is hard even for human beings and there is no clear criterion to measure it. To tackle this problem, in this paper, we propose LUMix, which model…
We demonstrate that locally linear embedding (LLE) inherently admits some unwanted results when no regularization is used, even for cases in which regularization is not supposed to be needed in the original algorithm. The existence of one special type of result, which we call ``projection pattern'', is mathematically proved in the situation that an exact local linear relation is achieved in each neighborhood of the data. These special patterns as well as some other bizarre results that may occur in more general situations are shown by numerical examples on the Swiss roll with a hole embedded i…
We apply concepts from manifold regularization to develop new regularization techniques for training locally stable deep neural networks. Our regularizers are based on a sparsification of the graph Laplacian which holds with high probability when the data is sparse in high dimensions, as is common in deep learning. Empirically, our networks exhibit stability in a diverse set of perturbation models, including $\ell_2$, $\ell_\infty$, and Wasserstein-based perturbations; in particular, we achieve 40% adversarial accuracy on CIFAR-10 against an adaptive PGD attack using $\ell_\infty$ perturbation…
Contrastive Learning Meets Pseudo-label-assisted Mixup Augmentation: A Comprehensive Graph Representation Framework from Local to Global
School of Information Science and Technology, Beijing University of Technology（北京工业大学信息科学与技术学院）；The University of Sydney Business School, The University of Sydney（悉尼大学商学院）；College of Information Science and Technology, Beijing University of Chemical Technology（北京化工大学信息科学与技术学院）
In semi-supervised learning (SSL), a technique called consistency regularization (CR) achieves high performance. It has been proved that the diversity of data used in CR is extremely important to obtain a model with high discrimination performance by CR. We propose a new data augmentation (Gradient-based Data Augmentation (GDA)) that is deterministically calculated from the image pixel value gradient of the posterior probability distribution that is the model output. We aim to secure effective data diversity for CR by utilizing three types of GDA. On the other hand, it has been demonstrated th…
A machine learning model that generalizes well should obtain low errors on unseen test examples. Thus, if we know how to optimally perturb training examples to account for test examples, we may achieve better generalization performance. However, obtaining such perturbation is not possible in standard machine learning frameworks as the distribution of the test data is unknown. To tackle this challenge, we propose a novel regularization method, meta-dropout, which learns to perturb the latent features of training examples for generalization in a meta-learning framework. Specifically, we meta-lea…
Computer Vision Center - Univ. Autònoma de Barcelona（计算机视觉中心 - 巴塞罗那自治大学）；Visual Tagging Services（视觉标签服务公司）
Regularization is key for deep learning since it allows training more complex models while keeping lower levels of overfitting. However, the most prevalent regularizations do not leverage all the capacity of the models since they rely on reducing the effective number of parameters. Feature decorrelation is an alternative for using the full capacity of the models but the overfitting reduction margins are too narrow given the overhead it introduces. In this paper, we show that regularizing negatively correlated features is an obstacle for effective decorrelation and present OrthoReg, a novel reg…
Western Washington University（西华盛顿大学）；Pacific Northwest National Laboratory（太平洋西北国家实验室）
Backdoor data poisoning attacks have recently been demonstrated in computer vision research as a potential safety risk for machine learning (ML) systems. Traditional data poisoning attacks manipulate training data to induce unreliability of an ML model, whereas backdoor data poisoning attacks maintain system performance unless the ML model is presented with an input containing an embedded "trigger" that provides a predetermined response advantageous to the adversary. Our work builds upon prior backdoor data-poisoning research for ML image classifiers and systematically assesses different exper…
Several data analysis techniques employ similarity relationships between data points to uncover the intrinsic dimension and geometric structure of the underlying data-generating mechanism. In this paper we work under the model assumption that the data is made of random perturbations of feature vectors lying on a low-dimensional manifold. We study two questions: how to define the similarity relationship over noisy data points, and what is the resulting impact of the choice of similarity in the extraction of global geometric information from the underlying manifold. We provide concrete mathemati…
