Beyond one-time relabeling: A sample inspection framework for correcting noisy labels in both classification and regression tasks

Product image recognition with guidance learning and noisy supervision

Crowdsourcing has become the most appealing way to provide a plethora of labels at a low cost. Nevertheless, labels from amateur workers are often noisy, which inevitably degenerates the robustness of subsequent learning models. To improve the label quality for subsequent use, majority voting (MV) is widely leveraged to aggregate crowdsourced labels due to its simplicity and scalability. However, when crowdsourced labels are "heavily" noisy (e.g., 40% of noisy labels), MV may not work well because of the fact "garbage (heavily noisy labels) in, garbage (full aggregated labels) out." This issue inspires us to think: if the ultimate target is to learn a robust model using noisy labels, why not provide partial aggregated labels and ensure that these labels are reliable enough for learning models? To solve this challenge by improving MV, we propose a coarse-to-fine label filtration model called double filter machine (DFM), which consists of a (majority) voting filter and a sparse filter serially. Specifically, the DFM refines crowdsourced labels from coarse filtering to fine filtering. In the stage of coarse filtering, the DFM aggregates crowdsourced labels by voting filter, which yields (quality-acceptable) full aggregated labels. In the stage of fine filtering, DFM further digs out a set of high-quality labels from full aggregated labels by sparse filter, since this filter can identify high-quality labels by the methodology of support selection. Based on the insight of compressed sensing, DFM recovers a ground-truth signal from heavily noisy data under a restricted isometry property. To sum up, the primary benefits of DFM are to keep the scalability by voting filter, while improve the robustness by sparse filter. We also derive theoretical guarantees for the convergence and recovery of DFM and reveal its complexity. We conduct comprehensive experiments on both the UCI simulated and the AMT crowdsourced datasets. Empirical results show that partial aggregated labels provided by DFM effectively improve the robustness of learning models.

In this paper a novel hybrid ensemble method aiming at the improvement of models accuracy in regression tasks is presented. The basic idea of the approach is the creation of an ensemble learner composed by a strong learner which is trained by exploiting data belonging to the whole training dataset and a set of specialised weak learners trained by using data coming from limited regions of the input space determined by means of a self organising map based clustering. In this context, different methods have been tested for the design of the learners, including a hierarchical approach. In the simulation phase, the strong and weak learners operate according to their punctual self-estimated reliabilities so as to exploit their strengths and overcome their weaknesses. The method has been tested on literature and real world datasets achieving competitive results by outperforming other ensemble methods on most of the tested datasets and reducing the average absolute error by up to 10%.

Leveraging weak or noisy supervision for building effective machine learning models has long been an important research problem. Its importance has further increased recently due to the growing need for large-scale datasets to train deep learning models. Weak or noisy supervision could originate from multiple sources including non-expert annotators or automatic labeling based on heuristics or user interaction signals. There is an extensive amount of previous work focusing on leveraging noisy labels. Most notably, recent work has shown impressive gains by using a meta-learned instance re-weighting approach where a meta-learning framework is used to assign instance weights to noisy labels. In this paper, we extend this approach via posing the problem as a label correction problem within a meta-learning framework. We view the label correction procedure as a meta-process and propose a new meta-learning based framework termed MLC (Meta Label Correction) for learning with noisy labels. Specifically, a label correction network is adopted as a meta-model to produce corrected labels for noisy labels while the main model is trained to leverage the corrected labels. Both models are jointly trained by solving a bi-level optimization problem. We run extensive experiments with different label noise levels and types on both image recognition and text classification tasks. We compare the re-weighing and correction approaches showing that the correction framing addresses some of the limitations of re-weighting. We also show that the proposed MLC approach outperforms previous methods in both image and language tasks.

<p>Extreme Learning Machine (ELM) is a training algorithm for Single-Layer Feed-forward Neural Network (SLFN). The difference in theory of ELM from other training algorithms is in the existence of explicitly-given solution due to the immutability of initialed weights. In practice, ELMs achieve performance similar to that of other state-of-the-art training techniques, while taking much less time to train a model. Experiments show that the speedup of training ELM is up to the 5 orders of magnitude comparing to standard Error Back-propagation algorithm.</p> <p>ELM is a recently discovered technique that has proved its efficiency in classic regression and classification tasks, including multi-class cases. In this thesis, extensions of ELMs for non-typical for Artificial Neural Networks (ANNs) problems are presented. The first extension, described in the third chapter, allows to use ELMs to get probabilistic outputs for multi-class classification problems. The standard way of solving this type of problems is based 'majority vote' of classifier's raw outputs. This approach can rise issues if the penalty for misclassification is different for different classes. In this case, having probability outputs would be more useful. In the scope of this extension, two methods are proposed. Additionally, an alternative way of interpreting probabilistic outputs is proposed.</p> <p>ELM method prove useful for non-linear dimensionality reduction and visualization, based on repetitive re-training and re-evaluation of model. The forth chapter introduces adaptations of ELM-based visualization for classification and regression tasks. A set of experiments has been conducted to prove that these adaptations provide better visualization results that can then be used for perform classification or regression on previously unseen samples.</p> <p>Shape registration of 3D models with non-isometric distortion is an open problem in 3D Computer Graphics and Computational Geometry. The fifth chapter discusses a novel approach for solving this problem by introducing a similarity metric for spectral descriptors. Practically, this approach has been implemented in two methods. The first one utilizes Siamese Neural Network to embed original spectral descriptors into a lower dimensional metric space, for which the Euclidean distance provides a good measure of similarity. The second method uses Extreme Learning Machines to learn similarity metric directly for original spectral descriptors. Over a set of experiments, the consistency of the proposed approach for solving deformable registration problem has been proven.</p>

With the rise of data engineering-driven automatic annotation strategies, deep learning has demonstrated remarkable performance and strong competitiveness in intelligent fault diagnosis. However, the inherent limitations of automatic annotators inevitably introduce noisy labels, which in turn hinder the generalization and accuracy of diagnostic models. Although numerous Learning with Noisy Labels (LNL) methods attempt to alleviate the impact of label noise through sample selection or label correction, most rely heavily on model predictions to guide training. This self-reinforcing mechanism frequently leads to confirmation bias, especially under high-noise conditions, thereby limiting their effectiveness. To address these challenges while preserving the full data utility, this paper proposes a novel approach termed the Multi-granularity Evidence Labels (MgEL), inspired by the principles of quantum entanglement and collapse. In MgEL, we perform feature-space fusion between entangled sub-distributions to construct a superposition state, from which two auxiliary labels are derived: a pseudo-label obtained by selecting the class with the maximum amplitude and a collapsed label sampled probabilistically according to the class-wise amplitude distribution. The collapsed label represents an uncertainty-aware observation, while the pseudo-label represents the most confident class estimation. These are then fused with the original annotation to form multi-granularity evidence labels. This approach allows MgEL to suppress confirmation bias and improve robustness under noisy supervision. Extensive experiments validate the effectiveness and reliability of MgEL, particularly in high-noise scenarios (e.g., noise intensity $\eta \ge 80\%$), underscoring its potential for practical deployment in low-cost, data-driven intelligent fault diagnosis systems.

Outlier-trimmed dual-interval smoothing loss for sample selection in learning with noisy labels.

In real application scenarios, the performance of deep networks may be degraded when the dataset contains noisy labels. Existing methods for learning with noisy labels are limited by two aspects. Firstly, methods based on the noise probability modeling can only be applied to class-level noisy labels. Secondly, others based on the memorization effect outperform in synthetic noise but get weak promotion in real-world noisy datasets. To solve these problems, this paper proposes a novel label-noise robust method named Discrepant Adversarial Training (DAT). The DAT method has ability of enforcing prominent feature extraction by matching feature distribution between clean and noisy data. Therefore, under the noise-free feature representation, the deep network can simply output the correct result. To better capture the divergence between the noisy and clean distribution, a new metric is designed to change the distribution divergence into computable. By minimizing the proposed metric with a min-max training of discrepancy on classifiers and generators, DAT can match noisy data to clean data in the feature space. To the best of our knowledge, DAT is the first to address the noisy label problem from the perspective of the feature distribution. Experiments on synthetic and real-world noisy datasets demonstrate that DAT can consistently outperform other state-of-the-art methods. Codes are available at https://github.com/Tyqnn0323/DAT.

Leveraging inexpensive and human intervention-based annotating methodologies, such as crowdsourcing and web crawling, often leads to datasets with noisy labels. Noisy labels can have a detrimental impact on the performance and generalization of deep neural networks. Robust models that are able to handle and mitigate the effect of these noisy labels are thus essential. In this work, we explore the open challenges of neural network memorization and uncertainty in creating robust learning algorithms with noisy labels. To overcome them, we propose a novel framework called “Bayesian DivideMix++” with two critical components: (i) DivideMix++, to enhance the robustness against memorization and (ii) Monte-Carlo MixMatch, which focuses on improving the effectiveness towards label uncertainty. DivideMix++ improves the pipeline by integrating the warm-up and augmentation pipeline with self-supervised pre-training and dedicated different data augmentations for loss analysis and backpropagation. Monte-Carlo MixMatch leverages uncertainty measurements to mitigate the influence of uncertain samples by reducing their weight in the data augmentation MixMatch step. We validate our proposed pipeline using four datasets encompassing various synthetic and real-world noise settings. We demonstrate the effectiveness and merits of our proposed pipeline using extensive experiments. Bayesian DivideMix++ outperforms the state-of-the-art models by considerable differences in all experiments. Our findings underscore the potential of leveraging these modifications to enhance the performance and generalization of deep neural networks in practical scenarios.

Supervised deep learning has achieved tremendous success in many computer vision tasks, which however is prone to overfit noisy labels. To mitigate the undesirable influence of noisy labels, robust loss functions offer a feasible approach to achieve noise-tolerant learning. In this work, we systematically study the problem of noise-tolerant learning with respect to both classification and regression. Specifically, we propose a new class of loss function, namely asymmetric loss functions (ALFs), which are tailored to satisfy the Bayes-optimal condition and thus are robust to noisy labels. For classification, we investigate general theoretical properties of ALFs on categorical noisy labels, and introduce the asymmetry ratio to measure the asymmetry of a loss function. We extend several commonly-used loss functions, and establish the necessary and sufficient conditions to make them asymmetric and thus noise-tolerant. For regression, we extend the concept of noise-tolerant learning for image restoration with continuous noisy labels. We theoretically prove that lp loss ( ) is noise-tolerant for targets with the additive white Gaussian noise. For targets with general noise, we introduce two losses as surrogates of l0 loss that seeks the mode when clean pixels keep dominant. Experimental results demonstrate that ALFs can achieve better or comparative performance compared with the state-of-the-arts.

This paper focuses on weakly supervised image understanding, in which the semantic labels are available only at image-level, without the specific object or scene location in an image. Existing algorithms implicitly assume that image-level labels are error-free, which might be too restrictive. In practice, image labels obtained from the pretrained predictors are easily contaminated. To solve this problem, we propose a novel algorithm for weakly supervised segmentation when only noisy image labels are available during training. More specifically, a semantic space is constructed first by encoding image labels through a graphlet (i.e., superpixel cluster) embedding process. Then, we observe that in the semantic space, the distribution of graphlets from images with a same label remains stable, regardless of the noises in image labels. Therefore, we propose a generative model, called latent stability analysis, to discover the stable patterns from images with noisy labels. Inferring graphlet semantics by making use of these mid-level stable patterns is much more secure and accurate than directly transferring noisy image-level labels into different regions. Finally, we calculate the semantics of each superpixel using maximum majority voting of its correlated graphlets. Comprehensive experimental results show that our algorithm performs impressively when the image labels are predicted by either the hand-crafted or deeply learned image descriptors.

The existence of noisy labels in real-world data negatively impacts the performance of deep learning models. Although much research effort has been devoted to improving the robustness towards noisy labels in classification tasks, the problem of noisy labels in deep metric learning (DML) remains under-explored. Existing noisy label learning methods designed for DML mainly discard suspicious noisy samples, resulting in a waste of the training data. To address this issue, we propose a noise-robust DML framework with SubGroup-based Positive-pair Selection (SGPS), which constructs reliable positive pairs for noisy samples to enhance the sample utilization. Specifically, SGPS first effectively identifies clean and noisy samples by a probability-based clean sample selectionstrategy. To further utilize the remaining noisy samples, we discover their potential similar samples based on the subgroup information given by a subgroup generation module and then aggregate them into informative positive prototypes for each noisy sample via a positive prototype generation module. Afterward, a new contrastive loss is tailored for the noisy samples with their selected positive pairs. SGPS can be easily integrated into the training process of existing pair-wise DML tasks, like image retrieval and face recognition. Extensive experiments on multiple synthetic and real-world large-scale label noise datasets demonstrate the effectiveness of our proposed method. Without any bells and whistles, our SGPS framework outperforms the state-of-the-art noisy label DML methods.

Noisy label problems are inevitably in existence within medical image segmentation causing severe performance degradation. Previous segmentation methods for noisy label problems only utilize a single image while the potential of leveraging the correlation between images has been overlooked. Especially for video segmentation, adjacent frames contain rich contextual information beneficial in cognizing noisy labels. Based on two insights, we propose a Multi-Scale Temporal Feature Affinity Learning (MS-TFAL) framework to resolve noisy-labeled medical video segmentation issues. First, we argue the sequential prior of videos is an effective reference, i.e., pixel-level features from adjacent frames are close in distance for the same class and far in distance otherwise. Therefore, Temporal Feature Affinity Learning (TFAL) is devised to indicate possible noisy labels by evaluating the affinity between pixels in two adjacent frames. We also notice that the noise distribution exhibits considerable variations across video, image, and pixel levels. In this way, we introduce Multi-Scale Supervision (MSS) to supervise the network from three different perspectives by re-weighting and refining the samples. This design enables the network to concentrate on clean samples in a coarse-to-fine manner. Experiments with both synthetic and real-world label noise demonstrate that our method outperforms recent state-of-the-art robust segmentation approaches. Code is available at https://github.com/BeileiCui/MS-TFAL.

Learning with noisy labels is one of the most challenging tasks in semi-supervised learning, and it poses significant problems in various practical applications. In the network learning process, the noisy labels concealed in the training dataset are easy to remember, resulting in poor generalization performance. To overcome this problem, inspired by the correction ability of humans – “think and learn from the past,” an end-to-end dynamic correction framework against label noise called Reflective Learning (RL) is proposed. This solution incorporates valuable knowledge from the past network training process to assist in correcting noisy labels. Specifically, during network training, a dynamic iterative function is implemented to adaptively correct noisy labels by employing the network’s predictive distribution information of all training epochs. This dynamic iterative function takes the form of a Standard Normal Distribution function to effectively match the changes of noisy label correction information contained in the network’s predictive probabilities. The proposed method is general and applicable to any backbone network and different types of noise without auxiliary information. Experiments are conducted on datasets with synthetic and real-world label noise datasets, including CIFAR-10, CIFAR-100, Tiny-ImageNet, and Clothing1M. They demonstrate that the proposed method is superior to the state-of-the-art results.

Inspired by the biological nervous system, deep neural networks (DNNs) are able to achieve remarkable performance in various tasks. However, they struggle to handle label noise, which can poison the memorization effects of DNNs. Co-teaching-based methods are popular in learning with noisy labels. These methods cross-train two DNNs based on the small-loss criterion and employ a strategy using either "disagreement" or "consistency" to obtain the divergence of the two networks. However, these methods are sample-inefficient for generalization in noisy scenarios. In this paper, we propose CoDC, a novel Co-teaching-basedmethod for accurate learning with label noise via both Disagreement and Consistency strategies. Specifically, CoDC maintains disagreement at the feature level and consistency at the prediction level using a balanced loss function. Additionally, a weighted cross-entropy loss is proposed based on information derived from the historical training process. Moreover, the valuable knowledge involved in "large-loss" samples is further developed and utilized by assigning pseudo-labels. Comprehensive experiments were conducted on both synthetic and real-world noise and under various noise types. CoDC achieved 72.81% accuracy on the Clothing1M dataset and 76.96% (Top1) accuracy on the WebVision1.0 dataset. These superior results demonstrate the effectiveness and robustness of learning with noisy labels.