Graph Neural Networks (GNNs) have become powerful models for both node- and graph-level tasks. While node-level learning focuses on individual nodes and their local structures, graph-level learning encounters challenges in capturing the global properties of graphs. In this paper, we conduct a theoretical and experimental analysis of existing graph-level learning frameworks and find that these frameworks typically adopt a single-view perspective based solely on node degree, which limits their ability to capture comprehensive graph characteristics. To address these issues, we propose a multi-view approach that leverages different types of centrality measures to capture diverse aspects of graph structure. We design an attention-based mechanism to adaptively integrate these multiple views, and use it as a readout function to perform weighted summation of node embeddings, termed as Adaptive Centrality Readout (ACRead). ACRead demonstrates enhanced flexibility and effectiveness when integrated with various GNN architectures, outperforming state-of-the-art readout methods, including KerRead and Set Transformer. Additionally, this multi-view centrality approach can serve as a standalone graph-level learning framework without relying on GNNs, referred to as Adaptive Centrality-based Graph Learning (ACGL), which achieves competitive performance by effectively combining different centrality perspectives.

Industrial Internet of Things (IIoT) plays a crucial role in advancing smart manufacturing by connecting numerous devices, enabling data exchanges, and supporting industrial applications. Yet, the timely and proper scheduling of asynchronously generated Heterogeneous Computing Tasks (HCTs) in IIoT environments remains a significant challenge. In this article, we first introduce the representation and notation of such HCTs and define a computing network structure. We then propose an initial mathematical programming-based scheduling model aimed at minimizing HCT completion time. To make this model easy to solve, we reformulate it by using logical constraints and derive a constraint programming-based model, for which a feasibility-guaranteed solution algorithm is developed. This algorithm leverages two easily-verified propositions to either identify feasible solutions or demonstrate the infeasibility of the problem.Furthermore, we have proven a critical proposition that facilitates the development of a hybrid programming-based scheduling approach, effectively combining the strengths of both mathematical and constraint programming models. As demonstrated through extensive computational experiments, our proposed approach achieves an average reduction of 20% in HCT completion time in comparison with its existing peers. It consistently and timely provides the high-quality solutions that meet the required deadlines.

Distributed parameter systems (DPS) are widely presented in various industrial fields. Time/space separation-based methods have proven to be effective modeling schemes for DPS. However, the sparse sensing environments in practical industrial scenarios inevitably result in incomplete data, posing significant challenges to the implementation of traditional modeling methods. In addition, the nonstationary spatiotemporal dynamics of the system pose another challenge for modeling. In this article, a sparse information completion-based incremental learning approach is proposed for modeling the complex DPS. First, a sparse information completion module is designed to reconstruct the nonsensor data, which takes spatial coupling effects into account. Then, the spatial basis functions are incrementally constructed to capture the systematic spatial variation. Finally, the temporal learning model is also incrementally developed to track temporal dynamics. Two case studies of sparse sensing in industrial processes demonstrate the superiority of the proposed modeling approach.

In the field of autonomous driving, semantic segmentation is crucial for scene understanding. Currently, there are two main methods: camera-based and Lidar-based approaches. To address the issues of Lidar segmentation lacking texture features and image segmentation lacking distance information, this paper proposes a fusion of camera and Lidar to achieve 3D semantic segmentation. The method utilizes a dual-stream encoder-decoder network to process camera images and Lidar point cloud and incorporates a specially designed attention mechanism module for feature fusion. To avoid expensive manual annotation of 3D point clouds, the study also introduces a cross-dataset and cross-modal self-supervised training approach. Experimental results show a 2.4% improvement compared to the Lidar-only mode baseline results on the SemanticKITTI dataset and a 6% improvement on the nuScenes dataset.

This paper proposed a learning-enabled approach for adaptive spatiotemporal modeling of industrial distributed processes. Within the framework of Karhunen-Loéve (KL) separation, the spatial basis functions (SBF) are updated online in a forgetful learning mode to capture spatial dynamics. Then, the temporal model is also updated iteratively in a forgetting mode to adjust temporal dynamics. Finally, the predicted spatiotemporal state is obtained via Time/Space synthesis. This dual forgetting mechanism embedded in the model can adaptively track the spatiotemporal dynamic changes, thus achieving better modeling effects. The experimental validation of distributed thermal processes in battery operation demonstrates the modeling efficacy of the designed learning approach.

The optimization of therapeutic antibodies through traditional techniques, such as candidate screening via hybridoma or phage display, is resource-intensive and time-consuming. In recent years, computational and artificial intelligence-based methods have been actively developed to accelerate and improve the development of therapeutic antibodies. In this study, we developed an end-to-end sequence-based deep learning model, termed AttABseq, for the predictions of the antigen-antibody binding affinity changes connected with antibody mutations. AttABseq is a highly efficient and generic attention-based model by utilizing diverse antigen-antibody complex sequences as the input to predict the binding affinity changes of residue mutations. The assessment on the three benchmark datasets illustrates that AttABseq is 120% more accurate than other sequence-based models in terms of the Pearson correlation coefficient between the predicted and experimental binding affinity changes. Moreover, AttABseq also either outperforms or competes favorably with the structure-based approaches. Furthermore, AttABseq consistently demonstrates robust predictive capabilities across a diverse array of conditions, underscoring its remarkable capacity for generalization across a wide spectrum of antigen-antibody complexes. It imposes no constraints on the quantity of altered residues, rendering it particularly applicable in scenarios where crystallographic structures remain unavailable. The attention-based interpretability analysis indicates that the causal effects of point mutations on antibody-antigen binding affinity changes can be visualized at the residue level, which might assist automated antibody sequence optimization. We believe that AttABseq provides a fiercely competitive answer to therapeutic antibody optimization.

Protein loops play a critical role in the dynamics of proteins and are essential for numerous biological functions, and various computational approaches to loop modeling have been proposed over the past decades. However, a comprehensive understanding of the strengths and weaknesses of each method is lacking. In this work, we constructed two high-quality datasets (i.e. the General dataset and the CASP dataset) and systematically evaluated the accuracy and efficiency of 13 commonly used loop modeling approaches from the perspective of loop lengths, protein classes and residue types. The results indicate that the knowledge-based method FREAD generally outperforms the other tested programs in most cases, but encountered challenges when predicting loops longer than 15 and 30 residues on the CASP and General datasets, respectively. The ab initio method Rosetta NGK demonstrated exceptional modeling accuracy for short loops with four to eight residues and achieved the highest success rate on the CASP dataset. The well-known AlphaFold2 and RoseTTAFold require more resources for better performance, but they exhibit promise for predicting loops longer than 16 and 30 residues in the CASP and General datasets. These observations can provide valuable insights for selecting suitable methods for specific loop modeling tasks and contribute to future advancements in the field.

Distributed parameter systems (DPS) widely exist in many industrial processes. Traditional modeling methods are not suitable for complex DPS under time-varying boundary conditions. To handle dynamics at different scales in the spatial and temporal domains, a dual-scale incremental learning approach is proposed for the efficient modeling of the complex time-varying DPS. Under the space/time separation framework, spatial basis functions (SBF) are first designed and updated incrementally at a slow scale over a long period of time. Under the given SBF, the temporal model will be incrementally iterated in real time (fast scale). An optimal choice of the fast/slow ratio can further improve the modeling performance by better coordinating the dynamics at different scales. The experiments on the curing oven thermal process can demonstrate the effectiveness of the proposed method for modeling complex time-varying dynamics of DPS.

Cracking the entangling code of protein-ligand interaction (PLI) is of great importance to structure-based drug design and discovery. Different physical and biochemical representations can be used to describe PLI such as energy terms and interaction fingerprints, which can be analyzed by machine learning (ML) algorithms to create ML-based scoring functions (MLSFs). Here, we propose the ML-based PLI capturer (ML-PLIC), a web platform that automatically characterizes PLI and generates MLSFs to identify the potential binders of a specific protein target through virtual screening (VS). ML-PLIC comprises five modules, including Docking for ligand docking, Descriptors for PLI generation, Modeling for MLSF training, Screening for VS and Pipeline for the integration of the aforementioned functions. We validated the MLSFs constructed by ML-PLIC in three benchmark datasets (Directory of Useful Decoys-Enhanced, Active as Decoys and TocoDecoy), demonstrating accuracy outperforming traditional docking tools and competitive performance to the deep learning-based SF, and provided a case study of the Serine/threonine-protein kinase WEE1 in which MLSFs were developed by using the ML-based VS pipeline in ML-PLIC. Underpinning the latest version of ML-PLIC is a powerful platform that incorporates physical and biological knowledge about PLI, leveraging PLI characterization and MLSF generation into the design of structure-based VS pipeline. The ML-PLIC web platform is now freely available at http://cadd.zju.edu.cn/plic/.