The whole slide image (WSI) classification is often formulated as a multiple instance learning (MIL) problem. Since the positive tissue is only a small fraction of the gigapixel WSI, existing MIL methods intuitively focus on identifying salient instances via attention mechanisms. However, this leads to a bias towards easy-to-classify instances while neglecting hard-to-classify instances. Some literature has revealed that hard examples are beneficial for modeling a discriminative boundary accurately. By applying such an idea at the instance level, we elaborate a novel MIL framework with masked hard instance mining (MHIM-MIL), which uses a Siamese structure (Teacher-Student) with a consistency constraint to explore the potential hard instances. With several instance masking strategies based on attention scores, MHIM-MIL employs a momentum teacher to implicitly mine hard instances for training the student model, which can be any attention-based MIL model. This counter-intuitive strategy essentially enables the student to learn a better discriminating boundary. Moreover, the student is used to update the teacher with an exponential moving average (EMA), which in turn identifies new hard instances for subsequent training iterations and stabilizes the optimization. Experimental results on the CAMELYON-16 and TCGA Lung Cancer datasets demonstrate that MHIM-MIL outperforms other latest methods in terms of performance and training cost. The code is available at: https://github.com/DearCaat/MHIM-MIL.

Automatic pavement distress classification facilitates improving the efficiency of pavement maintenance and reducing the cost of labor and resources. A recently influential branch of this task divides the pavement image into patches and infers the patch labels for addressing these issues from the perspective of multi-instance learning. However, these methods neglect the correlation between patches and suffer from a low efficiency in the model optimization and inference. As a representative approach of vision Transformer, Swin Transformer is able to address both of these issues. It first provides a succinct and efficient framework for encoding the divided patches as visual tokens, then employs self-attention to model their relations. Built upon Swin Transformer, we present a novel vision Transformer named Pavement Image Classification Transformer (PicT) for pavement distress classification. In order to better exploit the discriminative information of pavement images at the patch level, the Patch Labeling Teacher is proposed to leverage a teacher model to dynamically generate pseudo labels of patches from image labels during each iteration, and guides the model to learn the discriminative features of patches via patch label inference in a weakly supervised manner. The broad classification head of Swin Transformer may dilute the discriminative features of distressed patches in the feature aggregation step due to the small distressed area ratio of the pavement image. To overcome this drawback, we present a Patch Refiner to cluster patches into different groups and only select the highest distress-risk group to yield a slim head for the final image classification. We evaluate our method on a large-scale bituminous pavement distress dataset named CQU-BPDD. Extensive results demonstrate the superiority of our method over baselines and also show that PicT outperforms the second-best performed model by a large margin of +2.4% in [email protected] on detection task, +3.9% in F1 on recognition task, and 1.8x throughput, while enjoying 7x faster training speed using the same computing resources. Our codes and models have been released on https://github.com/DearCaat/PicT.

Summary With the continuous increase of the network scale, the structure of the network has also become complicated. The original community discovery algorithm based on small‐scale static networks has been unable to meet our needs. In order to improve the quality of community division, community discovery algorithms based on multiple optimization functions have been proposed. These multiobjective algorithms have continued to increase in time complexity as the optimization functions increase. The time complexity of the multiobjective community discovery algorithm is reduced, and the particle swarm algorithm has higher efficiency and accuracy in solving multiobjective optimization (MOO) problems. Based on the above background, the purpose of this article is to study a multiobjective particle swarm community discovery algorithm based on representation learning. This article uses the network representation learning method for static network community discovery, and designs an improved multiobjective particle swarm‐based community discovery algorithm (MOPSO‐CD). This randomness effectively prevents the algorithm from falling into a local optimum. At the same time, combined with the MOO algorithm, all the Pareto optimal solution sets are retained to adjust the population to correct the lack of accuracy caused by the randomness of the algorithm. In addition, in order to improve the efficiency of the algorithm, this article introduces an efficient Pareto optimal solution set method. Compared with the traditional MOO strategy, the time complexity of the MOO process is O(n2) Reduced to O(nlogn). Through experimental analysis, MOPSO‐CD has higher efficiency and community discovery quality.