Metric learning aims to learn a distance metric such that semantically similar instances are pulled together while dissimilar instances are pushed away. Many existing methods consider maximizing or at least constraining a distance margin in the feature space that separates similar and dissimilar pairs of instances to guarantee their generalization ability. In this article, we advocate imposing an adversarial margin in the input space so as to improve the generalization and robustness of metric learning algorithms. We first show that the adversarial margin, defined as the distance between training instances and their closest adversarial examples in the input space, takes account of both the distance margin in the feature space and the correlation between the metric and triplet constraints. Next, to enhance robustness to instance perturbation, we propose to enlarge the adversarial margin through minimizing a derived novel loss function termed the perturbation loss. The proposed loss can be viewed as a data-dependent regularizer and easily plugged into any existing metric learning methods. Finally, we show that the enlarged margin is beneficial to the generalization ability by using the theoretical technique of algorithmic robustness. Experimental results on 16 datasets demonstrate the superiority of the proposed method over existing state-of-the-art methods in both discrimination accuracy and robustness against possible noise.

Self-attention models have achieved the state-of-the-art performance in sequential recommender systems by capturing the sequential dependencies among user–item interactions. However, they rely on adding positional embeddings to the item sequence to retain the sequential information, which may break the semantics of item embeddings due to the heterogeneity between these two types of embeddings. In addition, most existing works assume that such dependencies exist solely in the item embeddings, but neglect their existence among the item features. In our previous study, we proposed a novel sequential recommendation model, i.e., MLP4Rec, based on the recent advances of MLP-Mixer architectures, which is naturally sensitive to the order of items in a sequence because matrix elements related to different positions of a sequence will be given different weights in training. We developed a tri-directional fusion scheme to coherently capture sequential, cross-channel, and cross-feature correlations with linear computational complexity as well as much fewer model parameters than existing self-attention methods. However, the cascading mixer structure, the large number of normalization layers between different mixer layers, and the noise generated by these operations limit the efficiency of information extraction and the effectiveness of MLP4Rec. In this extended version, we propose a novel framework – SMLP4Rec for sequential recommendation to address the aforementioned issues. The new framework changes the flawed cascading structure to a parallel mode, and integrates normalization layers to minimize their impact on the model’s efficiency while maximizing their effectiveness. As a result, the training speed and prediction accuracy of SMLP4Rec are vastly improved in comparison to MLP4Rec. Extensive experimental results demonstrate that the proposed method is significantly superior to the state-of-the-art approaches. The implementation code is available online to ease reproducibility.

Nocardia is clinically rare but highly pathogenic in clinical practice. Due to the lack of Nocardia screening methods, Nocardia is often missed in diagnosis, leading to worsening the condition. Therefore, this paper proposes a Nocardia screening method based on neural networks, aiming at quick Nocardia detection in sputum specimens with low costs and thereby reducing the missed diagnosis rate. Firstly, sputum specimens were collected from patients who were infected with Nocardia, and a part of the specimens were mixed with new sputum specimens from patients without Nocardia infection to enhance the data diversity. Secondly, the specimens were converted into smears with Gram staining. Images were captured under a microscope and subsequently annotated by experts, creating two datasets. Thirdly, each dataset was divided into three subsets: the training set, the validation set and the test set. The training and validation sets were used for training networks, while the test set was used for evaluating the effeteness of the trained networks. Finally, a neural network model was trained on this dataset, with an image of Gram-stained sputum smear as input, this model determines the presence and locations of Nocardia instances within the image. After training, the detection network was evaluated on two datasets, resulting in classification accuracies of 97.3% and 98.3%, respectively. This network can identify Nocardia instances in about 24 milliseconds per image on a personal computer. The detection metrics of mAP50 on both datasets were 0.780 and 0.841, respectively. The Nocardia screening method can accurately and efficiently determine whether Nocardia exists in the images of Gram-stained sputum smears. Additionally, it can precisely locate the Nocardia instances, assisting doctors in confirming the presence of Nocardia.

Resolving conformational heterogeneity in cryo-electron microscopy (cryo-EM) datasets remains a significant challenge in structural biology. Previous methods have often been restricted to working exclusively on volumetric densities, neglecting the potential of incorporating any pre-existing structural knowledge as prior or constraints. In this paper, we present a novel methodology, cryoSTAR, that harnesses atomic model information as structural regularization to elucidate such heterogeneity. Our method uniquely outputs both coarse-grained models and density maps, showcasing the molecular conformational changes at different levels. Validated against four diverse experimental datasets, spanning large complexes, a membrane protein, and a small single-chain protein, our results consistently demonstrate an efficient and effective solution to conformational heterogeneity with minimal human bias. By integrating atomic model insights with cryo-EM data, cryoSTAR represents a meaningful step forward, paving the way for a deeper understanding of dynamic biological processes.