Suppose we have a sample of instances paired with binary labels corrupted by arbitrary instance- and label-dependent noise. With sufficiently many such samples, can we optimally classify and rank instances with respect to the noise-free distribution? We provide a theoretical analysis of this question, with three main contributions. First, we prove that for instance-dependent noise, any algorithm that is consistent for classification on the noisy distribution is also consistent on the clean distribution. Second, we prove that for a broad class of instance- and label-dependent noise, a similar consistency result holds for the area under the ROC curve. Third, for the latter noise model, when the noise-free class-probability function belongs to the generalised linear model family, we show that the Isotron can efficiently and provably learn from the corrupted sample.
Effective and Robust Adversarial Training against Data and Label Corruptions
School of Information Technology & Electrical Engineering, the University of Queensland（昆士兰大学信息技术与电气工程学院）；School of Software, Shandong University（山东大学软件学院）
We conduct theoretical studies on streaming-based active learning for binary classification under unknown adversarial label corruptions. In this setting, every time before the learner observes a sample, the adversary decides whether to corrupt the label or not. First, we show that, in a benign corruption setting (which includes the misspecification setting as a special case), with a slight enlargement on the hypothesis elimination threshold, the classical RobustCAL framework can (surprisingly) achieve nearly the same label complexity guarantee as in the non-corrupted setting. However, this alg…
Centre for Artificial Intelligence, University of Technology Sydney（悉尼科技大学人工智能中心）
Deep neural networks (DNNs) have great expressive power, which can even memorize samples with wrong labels. It is vitally important to reiterate robustness and generalization in DNNs against label corruption. To this end, this paper studies the 0-1 loss, which has a monotonic relationship with an empirical adversary (reweighted) risk~\citep{hu2016does}. Although the 0-1 loss has some robust properties, it is difficult to optimize. To efficiently optimize the 0-1 loss while keeping its robust properties, we propose a very simple and efficient loss, i.e. curriculum loss (CL). Our CL is a tighter…
Columbia University（哥伦比亚大学）；Imperial College London（帝国理工学院）；The University of Alabama（阿拉巴马大学）
Supervised deep learning performance is heavily tied to the availability of high-quality labels for training. Neural networks can gradually overfit corrupted labels if directly trained on noisy datasets, leading to severe performance degradation at test time. In this paper, we propose a novel deep learning framework, namely Co-Seg, to collaboratively train segmentation networks on datasets which include low-quality noisy labels. Our approach first trains two networks simultaneously to sift through all samples and obtain a subset with reliable labels. Then, an efficient yet easily-implemented l…
We study the problem of learning generalized linear models under adversarial corruptions. We analyze a classical heuristic called the iterative trimmed maximum likelihood estimator which is known to be effective against label corruptions in practice. Under label corruptions, we prove that this simple estimator achieves minimax near-optimal risk on a wide range of generalized linear models, including Gaussian regression, Poisson regression and Binomial regression. Finally, we extend the estimator to the more challenging setting of label and covariate corruptions and demonstrate its robustness a…
Mohamed Bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）；Indian Institute of Technology Kanpur（印度理工学院坎普尔分校）
This paper presents SVAM (Sequential Variance-Altered MLE), a unified framework for learning generalized linear models under adversarial label corruption in training data. SVAM extends to tasks such as least squares regression, logistic regression, and gamma regression, whereas many existing works on learning with label corruptions focus only on least squares regression. SVAM is based on a novel variance reduction technique that may be of independent interest and works by iteratively solving weighted MLEs over variance-altered versions of the GLM objective. SVAM offers provable model recovery …
Semi-supervised learning (SSL) is the branch of machine learning that aims to improve learning performance by leveraging unlabeled data when labels are insufficient. Recently, SSL with deep models has proven to be successful on standard benchmark tasks. However, they are still vulnerable to various robustness threats in real-world applications as these benchmarks provide perfect unlabeled data, while in realistic scenarios, unlabeled data could be corrupted. Many researchers have pointed out that after exploiting corrupted unlabeled data, SSL suffers severe performance degradation problems. Th…
ETH Zürich（苏黎世联邦理工学院）；National University of Singapore（新加坡国立大学）
We consider the problem of optimizing an unknown (typically non-convex) function with a bounded norm in some Reproducing Kernel Hilbert Space (RKHS), based on noisy bandit feedback. We consider a novel variant of this problem in which the point evaluations are not only corrupted by random noise, but also adversarial corruptions. We introduce an algorithm Fast-Slow GP-UCB based on Gaussian process methods, randomized selection between two instances labeled "fast" (but non-robust) and "slow" (but robust), enlarged confidence bounds, and the principle of optimism under uncertainty. We present a n…
Successful training of deep neural networks with noisy labels is an essential capability as most real-world datasets contain some amount of mislabeled data. Left unmitigated, label noise can sharply degrade typical supervised learning approaches. In this paper, we present robust temporal ensembling (RTE), which combines robust loss with semi-supervised regularization methods to achieve noise-robust learning. We demonstrate that RTE achieves state-of-the-art performance across the CIFAR-10, CIFAR-100, ImageNet, WebVision, and Food-101N datasets, while forgoing the recent trend of label filterin…
Bilkent University（比尔肯特大学）；Aselsan（阿瑟尔桑）；Middle East Technical University（中东技术大学）
We introduce a comprehensive and statistical framework in a model free setting for a complete treatment of localized data corruptions due to severe noise sources, e.g., an occluder in the case of a visual recording. Within this framework, we propose i) a novel algorithm to efficiently separate, i.e., detect and localize, possible corruptions from a given suspicious data instance and ii) a Maximum A Posteriori (MAP) estimator to impute the corrupted data. As a generalization to Euclidean distance, we also propose a novel distance measure, which is based on the ranked deviations among the data a…
Weakly supervised machine learning algorithms are able to learn from ambiguous samples or labels, e.g., multi-instance learning or partial-label learning. However, in some real-world tasks, each training sample is associated with not only multiple instances but also a candidate label set that contains one ground-truth label and some false positive labels. Specifically, at least one instance pertains to the ground-truth label while no instance belongs to the false positive labels. In this paper, we formalize such problems as multi-instance partial-label learning (MIPL). Existing multi-instance …
The University of Tokyo（东京大学）；RIKEN AIP（理化学研究所先进智能项目中心）
Many weakly supervised classification methods employ a noise transition matrix to capture the class-conditional label corruption. To estimate the transition matrix from noisy data, existing methods often need to estimate the noisy class-posterior, which could be unreliable due to the overconfidence of neural networks. In this work, we propose a theoretically grounded method that can estimate the noise transition matrix and learn a classifier simultaneously, without relying on the error-prone noisy class-posterior estimation. Concretely, inspired by the characteristics of the stochastic label c…
Multi-instance learning (MIL) deals with objects represented as bags of instances and can predict instance labels from bag-level supervision. However, significant performance gaps exist between instance-level MIL algorithms and supervised learners since the instance labels are unavailable in MIL. Most existing MIL algorithms tackle the problem by treating multi-instance bags as harmful ambiguities and predicting instance labels by reducing the supervision inexactness. This work studies MIL from a new perspective by considering bags as auxiliary information, and utilize it to identify instance-…
Exploiting Conjugate Label Information for Multi-Instance Partial-Label Learning
School of Computer Science and Engineering, Southeast University（东南大学计算机科学与工程学院）；Key Lab. of Computer Network and Information Integration (Southeast University), MoE, China（教育部计算机网络与信息集成重点实验室（东南大学））；School of Information and Physical Sciences, The University of Newcastle（纽卡斯尔大学信息与物理科学学院）
Learning from Label Proportions (LLP) is a weakly supervised learning method that aims to perform instance classification from training data consisting of pairs of bags containing multiple instances and the class label proportions within the bags. Previous studies on multiclass LLP can be divided into two categories according to the learning task: per-instance label classification and per-bag label proportion estimation. However, these methods often results in high variance estimates of the risk when applied to complex models, or lack statistical learning theory arguments. To address this issu…
In multiple instance multiple label learning, each sample, a bag, consists of multiple instances. To alleviate labeling complexity, each sample is associated with a set of bag-level labels leaving instances within the bag unlabeled. This setting is more convenient and natural for representing complicated objects, which have multiple semantic meanings. Compared to single instance labeling, this approach allows for labeling larger datasets at an equivalent labeling cost. However, for sufficiently large datasets, labeling all bags may become prohibitively costly. Active learning uses an iterative…
Using Trusted Data to Train Deep Networks on Labels Corrupted by Severe Noise
University of California, Berkeley（加州大学伯克利分校）；University of Chicago（芝加哥大学）；Foundational Research Institute（基础研究所）；Toyota Technological Institute at Chicago（芝加哥丰田理工学院）
College of Software Engineering, Southeast University（东南大学软件学院）；School of Computer Science and Engineering, Southeast University（东南大学计算机科学与工程学院）；Yam Pak Charitable Foundation School of Computing and Information Sciences, Saint Francis University（圣弗朗西斯大学叶荫堂计算与信息科学学院）；Department of Computer Science, City University of Hong Kong（香港城市大学计算机科学系）
Information Sciences Institute（信息科学研究所）；University of Southern California（南加州大学）
Federated Learning has emerged as a dominant computational paradigm for distributed machine learning. Its unique data privacy properties allow us to collaboratively train models while offering participating clients certain privacy-preserving guarantees. However, in real-world applications, a federated environment may consist of a mixture of benevolent and malicious clients, with the latter aiming to corrupt and degrade federated model's performance. Different corruption schemes may be applied such as model poisoning and data corruption. Here, we focus on the latter, the susceptibility of feder…
Partial label learning (PLL) is a typical weakly supervised learning problem, where each training example is associated with a set of candidate labels among which only one is true. Most existing PLL approaches assume that the incorrect labels in each training example are randomly picked as the candidate labels and model the generation process of the candidate labels in a simple way. However, these approaches usually do not perform as well as expected due to the fact that the generation process of the candidate labels is always instance-dependent. Therefore, it deserves to be modeled in a refin…
Partial label learning (PLL) is a typical weakly supervised learning problem, where each training example is associated with a set of candidate labels among which only one is true. Most existing PLL approaches assume that the incorrect labels in each training example are randomly picked as the candidate labels. However, this assumption is not realistic since the candidate labels are always instance-dependent. In this paper, we consider instance-dependent PLL and assume that each example is associated with a latent label distribution constituted by the real number of each label, representing th…
In learning from label proportions (LLP), the instances are grouped into bags, and the task is to learn an instance classifier given relative class proportions in training bags. LLP is useful when obtaining individual instance labels is impossible or costly. In this work, we focus on the case of small bags, which allows to design an algorithm that explicitly considers all consistent instance label combinations. In particular, we propose an EM algorithm alternating between optimizing a general neural network instance classifier and incorporating bag-level annotations. Using two different image …
Department of Physics & Institute for Physics of Intelligence, University of Tokyo（东京大学物理与智能物理研究所）
Learning in the presence of label noise is a challenging yet important task: it is crucial to design models that are robust in the presence of mislabeled datasets. In this paper, we discover that a new class of loss functions called the gambler's loss provides strong robustness to label noise across various levels of corruption. We show that training with this loss function encourages the model to "abstain" from learning on the data points with noisy labels, resulting in a simple and effective method to improve robustness and generalization. In addition, we propose two practical extensions of …
The University of Sydney（悉尼大学）；Hong Kong Baptist University（香港浸会大学）；The University of Melbourne（墨尔本大学）；University of Science and Technology of China（中国科学技术大学）；RIKEN（理化学研究所）；The University of Tokyo（东京大学）
The problem of open-set noisy labels denotes that part of training data have a different label space that does not contain the true class. Lots of approaches, e.g., loss correction and label correction, cannot handle such open-set noisy labels well, since they need training data and test data to share the same label space, which does not hold for learning with open-set noisy labels. The state-of-the-art methods thus employ the sample selection approach to handle open-set noisy labels, which tries to select clean data from noisy data for network parameters updates. The discarded data are seen t…
Label noise is ubiquitous in various machine learning scenarios such as self-labeling with model predictions and erroneous data annotation. Many existing approaches are based on heuristics such as sample losses, which might not be flexible enough to achieve optimal solutions. Meta learning based methods address this issue by learning a data selection function, but can be hard to optimize. In light of these pros and cons, we propose Selection-Enhanced Noisy label Training (SENT) that does not rely on meta learning while having the flexibility of being data-driven. SENT transfers the noise distr…
Self-distillation (SD) is the process of first training a \enquote{teacher} model and then using its predictions to train a \enquote{student} model with the \textit{same} architecture. Specifically, the student's objective function is $\big(ξ*\ell(\text{teacher's predictions}, \text{ student's predictions}) + (1-ξ)*\ell(\text{given labels}, \text{ student's predictions})\big)$, where $\ell$ is some loss function and $ξ$ is some parameter $\in [0,1]$. Empirically, SD has been observed to provide performance gains in several settings. In this paper, we theoretically characterize the effect of SD…
Multi-instance learning is a type of weakly supervised learning. It deals with tasks where the data is a set of bags and each bag is a set of instances. Only the bag labels are observed whereas the labels for the instances are unknown. An important advantage of multi-instance learning is that by representing objects as a bag of instances, it is able to preserve the inherent dependencies among parts of the objects. Unfortunately, most existing algorithms assume all instances to be \textit{identically and independently distributed}, which violates real-world scenarios since the instances within …
Institute of Computational Perception, Johannes Kepler University Linz（约翰内斯·开普勒林茨大学计算感知研究所）
We explore how much can be learned from noisy labels in audio music tagging. Our experiments show that carefully annotated labels result in highest figures of merit, but even high amounts of noisy labels contain enough information for successful learning. Artificial corruption of curated data allows us to quantize this contribution of noisy labels.
University of Sydney（悉尼大学）；Xidian University（西安电子科技大学）；Hong Kong Baptist University（香港浸会大学）；University of Melbourne（墨尔本大学）；Brain-Inspired Technology Co., Ltd（脑启科技有限公司）；RIKEN（理化学研究所）；University of Tokyo（东京大学）
Learning with the \textit{instance-dependent} label noise is challenging, because it is hard to model such real-world noise. Note that there are psychological and physiological evidences showing that we humans perceive instances by decomposing them into parts. Annotators are therefore more likely to annotate instances based on the parts rather than the whole instances, where a wrong mapping from parts to classes may cause the instance-dependent label noise. Motivated by this human cognition, in this paper, we approximate the instance-dependent label noise by exploiting \textit{part-dependent} …
