Cluster and Scatter: A Multi-grained Active Semi-supervised Learning Framework for Scalable Person Re-identification

Two-phase self-supervised pretraining for object re-identification

Limited experimental data remains a key challenge in applying machine learning to drug discovery, particularly for cancer-related targets. In this study, we present a data-efficient active meta-deep learning framework to predict mitogen-activated protein kinase 1 (MAPK1) inhibitors, which are promising candidates for cancer-related therapies. Our approach integrates active learning (AL) with a meta-model that combines four deep architectures: a convolutional neural network, an attention, a graph convolutional network, and a graph neural network-attention, trained on molecular descriptors and graph-based representations. These models generate four probability-based features that feed into an attention-based meta-learner, improving predictive performance by 5.12% in the area under the precision-recall curve (AUPRC) and 5.48% in the Matthews correlation coefficient (MCC) using only 10% of the training data. Among the AL sampling strategies evaluated, entropy sampling showed competitive performance in selecting informative molecules for model improvement. Overall, our framework achieves an AUPRC of 0.835 ± 0.017 and MCC of 0.817 ± 0.017, on par with a traditional training method despite using only 26.7% of the training data. Compared to a conventional random forest model trained on brute-force, a 100% full training set, our approach shows a 10.6% improvement in AUPRC and modest gains in MCC, confirming the effectiveness of the proposed framework. Under severe class imbalance, balanced accuracy steadily increased across AL iterations, reaching values greater than 0.85 at the final iteration for all uncertainty-driven strategies. Molecular docking confirmed successful prioritization of the top four predicted compounds. Evaluation on an external MAPK1 data set demonstrated generalizability, with our approach achieving an AUPRC of 0.818 and an MCC of 0.403, comparable to the independent test set. These results highlight the potential of combining intelligent data selection with deep learning architectures through the meta-model to accelerate predictive performance in data-scarce drug discovery. Scientific contribution: This study contributes a novel, data-efficient active meta-deep learning framework for predicting MAPK1 inhibitors, addressing the challenge of limited experimental data in a cancer-specific target. By integrating AL with a meta-model composed of four deep architectures, the approach significantly enhances the predictive performance using only a fraction of the training data. The framework achieves superior metrics compared to traditional training methods, highlighting its potential to accelerate drug discovery in data-scarce settings.

Active learning is an important technology to solve the lack of data in crack detection model training. However, the sampling strategies of most existing active learning methods for crack detection are based on the uncertainty or representation of the samples, which cannot effectively balance the exploitation and exploration of active learning. To solve this problem, this study proposes an active learning method for crack detection based on subset searching and weighted sampling. First, a new active learning framework is established to successively search subsets with large uncertainty from the candidate dataset, and select training samples with large diversity from the subsets to update the crack detection model. Second, to realize the active learning process, a subset searching method based on sample relative error is proposed to adaptively select subsets with large uncertainty, and a weighted sampling method based on flow-based deep generative network is introduced to select training samples with large diversity form the subsets. Third, a termination criterion for active learning directly based on the prediction accuracy of the trained model is proposed to adaptively determine the maximum number of iterations. Finally, the proposed method is tested using two open-source crack datasets. The experimental comparison results on the Bridge Crack Library dataset show that the proposed method has higher calculation efficiency and prediction accuracy in crack detection than the uncertainty-based and representation-based active learning methods. The test results on the DeepCrack dataset show that the crack detection model trained by the proposed method has good transferability on different datasets with multi-scale concrete cracks and scenes.

Recently, Time Series Remote Sensing Images (TSRSIs) have been proven to be a significant resource for land use/land cover (LULC) mapping. Deep learning methods perform well in managing and processing temporal dependencies and have shown remarkable advancements within this domain. Although deep learning methods have exhibited outstanding performance in classifying TSRSIs, they rely on enough labeled time series samples for effective training. Labeling data with a wide geographical range and a long time span is highly time-consuming and labor-intensive. Active learning (AL) is a promising method of selecting the most informative data for labeling to save human labeling efforts. It has been widely applied in the remote sensing community, except for the classification of TSRSIs. The main challenge of AL in TSRSI classification is dealing with the internal temporal dependencies within TSRSIs and evaluating the informativeness of unlabeled time series data. In this paper, we propose a data-driven active deep learning framework for TSRSI classification to address the problem of limited labeled time series samples. First, a temporal classifier for TSRSI classification tasks is designed. Next, we propose an effective active learning method to select informative time series samples for labeling, which considers representativeness and uncertainty. For representativeness, we use the K-shape method to cluster time series data. For uncertainty, we construct an auxiliary deep network to evaluate the uncertainty of unlabeled data. The features with rich temporal information in the classifier's middle-hidden layers will be fed into the auxiliary deep network. Then, we define a new loss function with the aim of improving the deep model's performance. Finally, the proposed method in this paper was verified on two TSRSI datasets. The results demonstrate a significant advantage of our method over other approaches to TSRSI. On the MUDS dataset, when the initial number of samples was 100 after our method selected and labeled 2000 samples, an accuracy improvement of 4.92% was achieved. On the DynamicEarthNet dataset, when the initial number of samples was 1000 after our method selected and labeled 2000 samples, an accuracy improvement of 7.81% was attained. On the PASTIS dataset, when the initial number of samples was 1000 after our method selected and labeled 2000 samples, an accuracy improvement of 4.89% was achieved. Our code is available in Data Availability Statement.

Automatic analysis methods of electrocardiograms (ECGs) usually required large-scale annotated training data, but the annotation process is extremely time-consuming. While semi-supervised learning can leverage unlabeled data, its performance depends heavily on the quality of the initial labeled subset. Active learning has been used to identify the most informative samples for annotation, but conventional approaches face three critical limitations: (1) dependency on manual intervention for iterative query design, (2) prohibitive computational costs during sample selection, and (3) limited compatibility with semi-supervised learning frameworks. To address these limitations, we proposed an Unsupervised Active Feature-selective Semi-Supervised Learning (UAFSSL) framework for ECG analysis, including an unsupervised feature selection-based active learning module and a semi-supervised learning module. UAFSSL captures latent data distributions via unsupervised feature extraction, selects diverse and representative samples using pseudo-label clustering, and integrates seamlessly with semi-supervised learning to eliminate human intervention. We validated our algorithm on an ECG waveform segmentation task and an atrial fibrillation detection task. In the waveform segmentation task, our method improved the F1-score for P-wave delineation by 2.4% compared to random sampling, using only 5% of labeled samples. For the atrial fibrillation detection task, we evaluated our method on both the AFDB and a 24-hour dataset collected from 500 atrial fibrillation patients. Using only 200 labeled samples for model training, our method achieved AUC improvements of 2.5% and 2.2% over random sampling in five-fold cross validation. This is the first study to integrate unsupervised active learning with semi-supervised learning for automatic ECG analysis, offering a robust, automated solution to reduce annotation costs while enhancing clinical applicability.

Active learning is a useful technique for tasks for which unlabeled data is abundant but manual labeling is expensive. One example of such a task is semantic role labeling (SRL), which relies heavily on labels from trained linguistic experts. One challenge in applying active learning algorithms for SRL is that the complete knowledge of the SRL model is often unavailable, against the common assumption that active learning methods are aware of the details of the underlying models. In this paper, we present an active learning framework for black-box SRL models (i.e., models whose details are unknown). In lieu of a query strategy based on model details, we propose a neural query strategy model that embeds both language and semantic information to automatically learn the query strategy from predictions of an SRL model alone. Our experimental results demonstrate the effectiveness of both this new active learning framework and the neural query strategy model.

Active Learning is a machine learning and data mining technique that selects the most informative samples for labeling and uses them as training data; it is especially useful when there are large amount of unlabeled data and labeling them is expensive. Recently, batch-mode active learning, where a set of samples are selected concurrently for labeling, based on their collective merit, has attracted a lot of attention. The objective of batch-mode active learning is to select a set of informative samples so that a classifier learned on these samples has good generalization performance on the unlabeled data. Most of the existing batch-mode active learning methodologies try to achieve this by selecting samples based on certain criteria. In this article we propose a novel criterion which achieves good generalization performance of a classifier by specifically selecting a set of query samples that minimize the difference in distribution between the labeled and the unlabeled data, after annotation. We explicitly measure this difference based on all candidate subsets of the unlabeled data and select the best subset. The proposed objective is an NP-hard integer programming optimization problem. We provide two optimization techniques to solve this problem. In the first one, the problem is transformed into a convex quadratic programming problem and in the second method the problem is transformed into a linear programming problem. Our empirical studies using publicly available UCI datasets and two biomedical image databases demonstrate the effectiveness of the proposed approach in comparison with the state-of-the-art batch-mode active learning methods. We also present two extensions of the proposed approach, which incorporate uncertainty of the predicted labels of the unlabeled data and transfer learning in the proposed formulation. In addition, we present a joint optimization framework for performing both transfer and active learning simultaneously unlike the existing approaches of learning in two separate stages, that is, typically, transfer learning followed by active learning. We specifically minimize a common objective of reducing distribution difference between the domain adapted source, the queried and labeled samples and the rest of the unlabeled target domain data. Our empirical studies on two biomedical image databases and on a publicly available 20 Newsgroups dataset show that incorporation of uncertainty information and transfer learning further improves the performance of the proposed active learning based classifier. Our empirical studies also show that the proposed transfer-active method based on the joint optimization framework performs significantly better than a framework which implements transfer and active learning in two separate stages.

Active learning (AL) is a well‐known technique to optimize data usage in training, through the interactive selection of unlabeled observations, out of a large pool of unlabeled data, to be labeled by a supervisor. Its focus is to find the unlabeled observations that, once labeled, will maximize the informativeness of the training dataset, therefore reducing data‐related costs. The literature describes several methods to improve the effectiveness of this process. Nonetheless, there is a paucity of research developed around the application of artificial data sources in AL, especially outside image classification or NLP. This paper proposes a new AL framework, which relies on the effective use of artificial data. It may be used with any classifier, generation mechanism, and data type and can be integrated with multiple other state‐of‐the‐art AL contributions. This combination is expected to increase the ML classifier’s performance and reduce both the supervisor’s involvement and the amount of required labeled data at the expense of a marginal increase in computational time. The proposed method introduces a hyperparameter optimization component to improve the generation of artificial instances during the AL process as well as an uncertainty‐based data generation mechanism. We compare the proposed method to the standard framework and an oversampling‐based active learning method for more informed data generation in an AL context. The models’ performance was tested using four different classifiers, two AL‐specific performance metrics, and three classification performance metrics over 15 different datasets. We demonstrated that the proposed framework, using data augmentation, significantly improved the performance of AL, both in terms of classification performance and data selection efficiency (all the codes and preprocessed data developed for this study are available at https://github.com/joaopfonseca/publications/).

Towards fast and kernelized orthogonal discriminant analysis on person re-identification

Developing intelligent learning algorithms under the constraint of limited manual labor is a fundamental research challenge and a problem of immense practical importance. Active learning algorithms alleviate this problem by automatically identifying the salient and exemplar instances from large amounts of unlabeled data; this tremendously reduces the human annotation effort as only a small subset of the samples, identi ed by the algorithm, needs to be labeled manually. Further, the unprecedented increase in the volume of digital data has necessitated the usage of multiple, independent computers for its storage and processing, in a given application. The need of the hour is therefore an active learning framework which can operate in a distributed setup, where the unlabeled data is partitioned across multiple computers. In this paper, we propose a novel algorithm to address this important challenge. Our algorithm requires minimal communication among the computers (over which the data is stored) and also enjoys nice theoretical properties. Our extensive empirical studies on a variety of challenging, real-world vision datasets, from different application domains, corroborate the potential of the proposed framework.

Data-driven Fault Detection and Diagnostics (FDD) methods often assume that sufficient labeled samples are class-balanced and faulty classes in testing are precedent or seen previously during model training. When monitoring a large fleet of assets at scale, these assumptions may be violated: (I) only a limited number of samples can be manually labeled due to constraints of time and/or cost; (II) most of the samples collected in the engineering systems are under normal conditions, leading to a highly imbalanced class distribution and a biased prediction model. This work presents a robust and cost-effective FDD framework that integrates active learning and semi-supervised learning methods to detect both known and unknown failure modes iteratively. This framework allows to strategically select the samples to be annotated from a fully unlabeled dataset, while labeling cost is minimal. Specifically, a novel graph-based semi-supervised classifier with adaptive graph construction is developed to predict labels with imbalanced data and detect novel classes. We designed a multi-criteria active learning sampling strategy to select the most informative samples from unlabeled data in order to query minimal number of labels for classification. We tested the framework and algorithms in three synthetic datasets and one real-world dataset of vehicle air intake systems, and demonstrated the superior performance compared to the state-of-the-art methods for fleet-level FDD.

Active semi-supervised learning with multiple complementary information

Deep learning has shown significant successes in person reidentification (re-id) tasks. However, most existing works focus on discriminative feature learning and impose complex neural networks, suffering from low inference efficiency. In fact, feature extraction time is also crucial for real-world applications and lightweight models are needed. Prevailing pruning methods usually pay attention to compact classification models. However, these methods are suboptimal for compacting re-id models, which usually produce continuous features and are sensitive to network pruning. The key point of pruning re-id models is how to retain the original filter distribution in continuous features as much as possible. In this work, we propose a blockwise adjacent filter decaying method to fill this gap. Specifically, given a trained model, we first evaluate the redundancy of filters based on the adjacency relationships to preserve the original filter distribution. Second, previous layerwise pruning methods ignore that discriminative information is enhanced block-by-block. Therefore, we propose a blockwise filter pruning strategy to better utilize the block relations in the pretrained model. Third, we propose a novel filter decaying policy to progressively reduce the scale of redundant filters. Different from conventional soft filter pruning that directly sets the filter values as zeros, the proposed filter decaying can keep the pretrained knowledge as much as possible. We evaluate our method on three popular person reidentification datasets, that is: 1) Market-1501; 2) DukeMTMC-reID; and 3) MSMT17_V1. The proposed method shows superior performance to the existing state-of-the-art pruning methods. After pruning over 91.9% parameters on DukeMTMC-reID, the Rank-1 accuracy only drops 3.7%, demonstrating its effectiveness for compacting person reidentification.

Hemmer, Patrick Kühl, Niklas Schöffer, Jakob Convolutional Neural Networks (CNNs) have proven to be state-of-the-art models for supervised computer vision tasks, such as image classification. However, large labeled datasets are generally needed for the training and validation of such models. In many domains, unlabeled data is available but labeling is expensive, for instance, when specific expert knowledge is required. Active Learning (AL) is one approach to mitigate the problem of limited labeled data. Through selecting the most informative and representative data instances for labeling, AL can contribute to more efficient learning of a model. Recent AL methods for CNNs propose different solutions for the selection of to-be-labeled instances. However, they do not perform consistently well and are often computationally expensive. In this paper, we propose a novel AL algorithm that efficiently learns from unlabeled data by capturing high prediction uncertainty. By replacing the softmax standard output of a CNN with the parameters of a Dirichlet density, the model learns to identify data instances that contribute efficiently to improving model performance during training. We demonstrate in several experiments with publicly available data that our method consistently outperforms other state-of-the-art AL approaches. It can be easily implemented and does not require extensive computational resources for training. Additionally, we are able to show the benefits of the approach on the real-world medical use case of pneumonia detection in chest X-ray images.

Abstract. Person Re-Identification (Re-ID) is the task of matching the same person in multiple images captured by different cameras. Recently deep learning-based Re-ID algorithms demonstrated an exciting progress for terrestrial-based cameras, still, person Re-ID in aerial images poses multiple challenges including occlusion of human feature parts, image distortion, and dynamic camera location. In this paper, we propose a new Person Aerial Re-ID framework Robust to Occlusion and Thermal imagery (ParrotGAN). Our model is focused on cross-modality person Re-ID in aerial images. Furthermore, we collected a new large-scale synthetic multimodal AerialReID dataset with 30k images and 137 person ID. Our ParrotGAN model leverages two strategies to achieve robust performance in the task of person Re-ID in thermal and visible range. Firstly, we use a latent space of StyleGAN2 model to estimate the distance between two images of a person. Specifically, we project each real image into the latent space with a correspondent latent vector z. We use the distance between latent vectors to provide a Re-ID similarity metric. Secondly, we use a generative-adversarial network to translate a color image to a synthetic thermal image. We use synthetic image for a cross-modality Re-ID. We evaluate our ParrotGAN model and basleines on our AerialReID and PRAI-1581 datasets. The results of the evaluation are encouraging and demonstrate that our ParrotGAN model competes with baselines in visible range aerial person Re-ID and outperforms them in the cross-modality setting. We made our code and dataset publicly available.