Active learning (AL) is a semi-supervised learning paradigm with human-machine interaction and a limited annotation budget. However, few AL studies have explored distribution inconsistency between the data and the population. In this paper, we consider a basic form of the aforementioned issue,i.e., the training data is non-independently and identically distributed (non-IID) sampled from a class uniformly distributed population. Accordingly, we propose a naïve sample selection plugin, namely generalized active stratified sampling (GASS), to rebalance the sample size of each class during AL iterative process, resulting in a progressive approximation to the population. We generalize statistical stratified sampling to support the uncertainty strata criterion, forming the statistical foundation of GASS. This method, as a plugin, can seamlessly collaborate with popular information-based strategies. GASS shows superior rebalancing capabilities by analyzing the statistical moment and the class imbalanced index under the Probably Approximately Correct (PAC) theory. Furthermore, models derived with GASS have low Rademacher complexity (RC), indicating low generalization error bounds, and GASS also exhibits strong robustness to prediction perturbations. Experiments were conducted on 5 benchmark image datasets, and the results show that GASS significantly boosts the test accuracy by about$2.38\%$/$3.19\%$(paired$t$-test$p=0.01$/0.04) and reduces the empirical RC by about$1.42\%$/$1.94\%$(paired$t$-test$p=0.01$/0.05) on average in class imbalanced/balanced scenarios, respectively. This study establishes a potential benchmark for information-based AL.

In this research, we propose a novel denoising diffusion model based on shortest-path modeling that optimizes residual propagation to enhance both denoising efficiency and quality. Drawing on Denoising Diffusion Implicit Models (DDIM) and insights from graph theory, our model, termed the Shortest Path Diffusion Model (ShortDF), treats the denoising process as a shortest-path problem aimed at minimizing reconstruction error. By optimizing the initial residuals, we improve the efficiency of the reverse diffusion process and the quality of the generated samples. Extensive experiments on multiple standard benchmarks demonstrate that ShortDF significantly reduces diffusion time (or steps) while enhancing the visual fidelity of generated samples compared to prior arts. This work, we suppose, paves the way for interactive diffusion-based applications and establishes a foundation for rapid data generation. Code is available at https://github.com/UnicomAI/ShortDF.

Deep active learning (DeepAL) extends supervised deep learning to human-machine interactive scenarios with limited annotation budgets. Most existing DeepAL approaches for visual recognition fail to consider the intrinsic hierarchical structure and dependencies between labels. In this article, we propose a unified DeepAL framework for the aforementioned challenge, which fuses three tightly coupled techniques: 1) hierarchical dependency representation entropy (HDRE); 2) approximate class-balanced typical sampling (ACTS); and 3) local probability suppression loss. First, the HDRE provides the features of information entropy, interclass dependencies, and constraints effectively. It is used to determine the query priority of unlabeled samples. Second, the ACTS, embedded with the HDRE, is designed for querying, where the optimal sample query size of each class is derived. It excludes samples near the boundary by employing a well-designed hierarchical margin sampling. Third, the local probability suppression loss is a transfer-friendly loss function that enables the deep model to flatly fit data with a hierarchical structure. It compensates for hierarchical dependencies between classes using the local probability suppression constraint, modeling conditional and unconditional probabilities simultaneously. We conducted experiments on five public image datasets, and the results demonstrated the effectiveness of our approach.

Fine-grained visual categorization is a challenging issue owing to high intra-class and low inter-class variances. Classical approaches rely on pre-trained models or many fine annotations. In this paper, we observe that spatial and frequency information provides distinct image views, and propose a new spatial–frequency feature fusion (SFFF) perspective to handle this challenging issue. Specifically, we design a heterogeneous feature extraction loss function, construct a global and local fusion SFFF network, and propose an importance-sparsity selection strategy. For feature extraction, we focus on the frequency domain feature learning network, extract fine-grained features, and achieve feature complementarity. For feature selection, we propose importance ranking and sparse regularity to constrain spatial–frequency features. For feature fusion, we design a spatial–frequency loss and an inter-layer switching strategy to achieve local-global collaboration. Comparative experiments were performed on popular fine-grained datasets and classic datasets such as CUB200-2011, Stanford Cars, Stanford Dogs, FGVC-Aircraft, and CIFAR100. The effectiveness and outstanding performance of SFFF are confirmed by comparisons with more than 40 state-of-the-art fine-grained categorization methods. Ablation studies and visualizations are provided to facilitate an understanding of our approach.

Practical data often suffer from missing attribute values and lack of class labels. A reasonable machine learning scenario involves obtaining certain values and labels at cost on request. In this article, we propose the cost-sensitive active learning through unified evaluation and dynamic selection (CALS) algorithm to handle the learning task in this new scenario. For data representation, we consider misclassification cost, label query cost, and attribute query cost. For the cost/benefit estimation, we design a unified assessment of attribute values and labels with softmax regression. For the selection of attribute value and label, we propose an optimal acquisition scheme with permutation and greedy strategies. We perform experiments with synthetic, benchmark, and domain datasets. The results of the significance test verify the effectiveness of CALS and its superiority over cost-sensitive active learning and missing data imputation algorithms.

Clustering-based active learning approaches take advantage of the structure of the data to select representative instances. However, some algorithms are either inefficient or only applicable to some data. In this paper, we propose an effective and adaptive algorithm that will be called active learning through two-stage clustering (ALTA). The first stage is data preprocessing using the two-round-clustering algorithm to obtain $\sqrt n $ small blocks, where n is the number of instances. For each block, the closest instance of the center is selected as the representative. The second stage is the active learning of representative instances through density clustering. This stage consists of a number of iterations of density clustering, labeling and classification. In general, data preprocessing reduces the size of the data and the complexity of the algorithm. The combination of distance vector clustering and density clustering makes the algorithm more adaptive. Experiments are performed in comparison against the state-of-the-art active learning algorithms on nine datasets. Results demonstrate that the new algorithm has higher classification accuracy with the same number of labeled data.