Enhancing Coffee Leaf Disease Classification via Active Learning and Diverse Sample Selection

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.

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.

A structured perspective of volumes on active learning

The current exponential increase of spatiotemporally explicit data streams from satellitebased Earth observation missions offers promising opportunities for global vegetation monitoring. Intelligent sampling through active learning (AL) heuristics provides a pathway for fast inference of essential vegetation variables by means of hybrid retrieval approaches, i.e., machine learning regression algorithms trained by radiative transfer model (RTM) simulations. In this study we summarize AL theory and perform a brief systematic literature survey about AL heuristics used in the context of Earth observation regression problems over terrestrial targets. Across all relevant studies it appeared that: (i) retrieval accuracy of AL-optimized training data sets outperformed models trained over large randomly sampled data sets, and (ii) Euclidean distance-based (EBD) diversity method tends to be the most efficient AL technique in terms of accuracy and computational demand. Additionally, a case study is presented based on experimental data employing both uncertainty and diversity AL criteria. Hereby, a a simulated training data base by the PROSAIL-PRO canopy RTM is used to demonstrate the benefit of AL techniques for the estimation of total leaf carotenoid content (Cxc) and leaf water content (Cw). Gaussian process regression (GPR) was incorporated to minimize and optimize the training data set with AL. Training the GPR algorithm on optimally AL-based sampled data sets led to improved variable retrievals compared to training on full data pools, which is further demonstrated on a mapping example. From these findings we can recommend the use of AL-based sub-sampling procedures to select the most informative samples out of large training data pools. This will not only optimize regression accuracy due to exclusion of redundant information, but also speed up processing time and reduce final model size of kernel-based machine learning regression algorithms, such as GPR. With this study we want to encourage further testing and implementation of AL sampling methods for hybrid retrieval workflows. AL can contribute to the solution of regression problems within the framework of operational vegetation monitoring using satellite imaging spectroscopy data, and may strongly facilitate data processing for cloud-computing platforms.

Collection of training samples for remote sensing image classification is always time-consuming and expensive. In this context, active learning (AL) that aims at using limited training samples to achieve promising classification performances is developed. Recently, integration of spatial information into AL exhibits new potential for image classification. In this letter, an AL approach with two-stage spatial computation (AL-2SC) is proposed to improve the selection of training samples. The spatial features derived from remote sensing image and the probability outputs from the neighboring pixels are introduced in AL process. Moreover, we compare several AL approaches which take spatial information into account. In experiments, random sampling (RS) and four AL methods, including AL using breaking ties heuristic (BT), AL with spatial feature (AL-SF), AL with neighbouring responses (AL-NR), and AL-2SC, are considered. Three remote sensing datasets, including one hyperspectral and two multispectral images, are used to compare the performance of different methods. It is illustrated that, the utilization of spatial information is very important for the improvement of AL performance, and the proposed AL-2SC shows the most satisfactory result.

Neural networks have recently been shown to be highly effective at predicting links for constructing knowledge graphs. Existing research has mainly focused on designing 1) deep neural network models that are expressive in capturing fine-grained semantics, e.g., NTN and ConvE, but that are however less scalable; or 2) shallow models that are scalable, e.g., TransE and DistMult, yet limited in capturing expressive semantic features. In this work, we demonstrate that we can get the best of both worlds while drastically reducing the amount of data needed to train a deep network by leveraging active learning. We present a novel deep active learning framework, ActiveLink, which can be applied to actively train any neural link predictor. Inspired by recent advances in Bayesian deep learning, ActiveLink takes a Bayesian view on neural link predictors, thereby enabling uncertainty sampling for deep active learning. ActiveLink extends uncertainty sampling by exploiting the underlying structure of the knowledge graph, i.e., links between entities, to improve sampling effectiveness. To accelerate model training, ActiveLink further adopts an incremental training method that allows deep neural networks to be incrementally trained while optimizing their generalizability at each iteration. Extensive validation on real-world datasets shows that ActiveLink is able to match state-of-the-art approaches while requiring only 20% of the original training data.

The prediction of the ablation rate of silicone rubber-based composites is of great significance to accelerate the development of flexible thermal protection materials. Herein, a method which combines uniform design experimentation, active learning, and virtual sample generation was proposed to establish a prediction model of the mass ablation rate based on a small dataset. Briefly, a small number of sample points were collected using uniform design experimentation, which were marked to construct the initial dataset and primitive model. Then, data points were acquired from the sample pool and iterated using various integrated algorithms through active learning to update the above dataset and model. Finally, a large number of virtual samples were generated based on the optimal model, and a further optimized prediction model was achieved. The results showed that after introducing 300 virtual samples, the average percentage error of the gradient boosting decision tree (GBDT) prediction model on the test set decreased to 3.1%, which demonstrates the effectiveness of the proposed method in building prediction models based on a small dataset.

Compressive sampling (CS), or Compressed Sensing , has generated a tremendous amount of excitement in the signal processing community. Compressive sampling, which involves non-traditional samples in the form of randomized projections, can capture most of the salient information in a signal with a relatively small number of samples, often far fewer samples than required using traditional sampling schemes. Adaptive sampling (AS), also called Active Learning , uses information gleaned from previous observations (e.g. , feedback) to focus the sampling process. Theoretical and experimental results have shown that adaptive sampling can dramatically outperform conventional (non-adaptive) sampling schemes. This paper compares the theoretical performance of compressive and adaptive sampling for regression in noisy conditions, and it is shown that for certain classes of piecewise constant signals and high SNR regimes both CS and AS are near optimal. This result is remarkable since it is the first evidence that shows that compressive sampling, which is non-adaptive, cannot be significantly outperformed by any other method (including adaptive sampling procedures), even in the presence of noise. The performance of CS schemes for signal detection is also investigated.© (2006) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

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.

This study presents a novel approach to medical text classification using a deep active incremental learning model, aiming to improve the automation of the preauthorization process in medical health insurance. By automating decision-making for request approval or denial through text classification techniques, the primary focus is on real-time prediction, utilization of limited labeled data, and continuous model improvement. The proposed approach combines a Bidirectional Long Short-Term Memory (Bi-LSTM) neural network with active learning, using uncertainty sampling to facilitate expert-based sample selection and online learning for continuous updates. The proposed model demonstrates improved predictive accuracy over a baseline Long Short-Term Memory (LSTM) model. Through active learning iterations, the proposed model achieved a 4% improvement in balanced accuracy over 100 iterations, underscoring its efficiency in continuous refinement using limited labeled data.

Bio-inspired Active Learning method in spiking neural network

Active learning is a learning mechanism which can actively query the user for labels. The goal of an active learning algorithm is to build an effective training set by selecting those most informative samples and improve the efficiency of the model within the limited time and resource. In this paper, we mainly focus on a state-of-the-art active learning method, the SVM-based margin sampling. However, margin sampling does not consider the distribution and the structural space connectivity among the unlabeled data when several examples are chosen simultaneously, which may lead to oversampling on dense regions. To overcome this shortcoming, we propose an improved margin sampling method by applying the manifold-preserving graph reduction algorithm to the original margin sampling method. Experimental results on multiple data sets demonstrate that our method obtains better classification performance compared with the original margin sampling.

Video annotation is an effective way to facilitate content-based analysis for videos. Automatic machine learning methods are commonly used to accomplish this task. Among these, active learning is one of the most effective methods, especially when the training data cost a great deal to obtain. One of the most challenging problems in active learning is the sample selection. Various sampling strategies can be used, such as uncertainty, density, and diversity, but it is difficult to strike a balance among them. In this paper, we provide a visualization-based batch mode sampling method to handle such a problem. An iso-contour-based scatterplot is used to provide intuitive clues for the representativeness and informativeness of samples and assist users in sample selection. A semisupervised metric learning method is incorporated to help generate an effective scatterplot reflecting the high-level semantic similarity for visual sample selection. Moreover, both quantitative and qualitative evaluations are provided to show that the visualization-based method can effectively enhance sample selection in active learning.

Multiclass autoencoder-based active learning for sensor-based human activity recognition

Food security continues to be a significant challenge the world over, with crop production becoming increasingly threatened by crop diseases and pest infestations. In the case of chili production, farmers often suffer significant yield loss and economic insecurity due to the unpredictable nature of both of these problems. Current pest control options (agrochemical and organic methods alike) have not reliably been enough for timely and effective control and demonstrate the importance of early and effective pest and disease identification processes. To solve this problem, the present work describes a deep learning implementation that leverages advanced YOLO-based architectures for the detection and classification of leaf diseases and pest on chili plant leaves. The present work trains and evaluates three models (YOLOv5, YOLOv7, and YOLOv8) on a newly created and balanced dataset of over 28,800 images combining 20 total classes of pest and leaf diseases. The dataset was supplemented by preprocessing the images and conducting an augmentation process to create a total of 32,000 images for training to generate reliable models. The results from the experiments found that YOLOv8 provided the best baseline performance of 95.1% mean Average Precision (mAP), while YOLOv5 had an mAP of 86.1%, and YOLOv7 had an mAP of 67.5%. An additional enhancement in the construction of a Modified YOLOv8 hybrid model—reflecting all advantages of YOLOv5, YOLOv7, and YOLOv8—achieved a highest mAP of 99.5% to be the most effective model in this study, the results suggest that the newly proposed Modified YOLOv8 framework, is highly accurate and reliable for the early detection of pests and diseases in chili, and is helpful to improve sustainable agricultural practices, mitigate crop losses, and increase global food security.