This paper presents a dual-model assisted evolutionary algorithm designed to improve the performance of decomposition-based multi-objective evolutionary algorithm (MOEA/D) when dealing with expensive problems. Firstly, the algorithm utilises a dual-model management strategy, which replaces the real evaluation by constructing two models and generating deviation parameters through cross-validation. This enables the final prediction to be derived by combining the model predictions and deviation parameters. Secondly, a new optimization strategy was implemented whereby the same population of parents was selected in each generation and two candidate offspring solutions were generated using two reproduction operators, selecting the most promising as the offspring. Empirical results show that the method is competitive on a variety of benchmark problems. It can also efficiently solve expensive multi-objective optimization problems within a limited computational budget, and its performance exceeds that of other popular surrogate-assisted evolutionary algorithms.

MOTIVATION: Molecular property prediction is a significant requirement in AI-driven drug design and discovery, aiming to predict the molecular property information (e.g. toxicity) based on the mined biomolecular knowledge. Although graph neural networks have been proven powerful in predicting molecular property, unbalanced labeled data and poor generalization capability for new-synthesized molecules are always key issues that hinder further improvement of molecular encoding performance. RESULTS: We propose a novel self-supervised representation learning scheme based on a Cascaded Attention Network and Graph Contrastive Learning (CasANGCL). We design a new graph network variant, designated as cascaded attention network, to encode local-global molecular representations. We construct a two-stage contrast predictor framework to tackle the label imbalance problem of training molecular samples, which is an integrated end-to-end learning scheme. Moreover, we utilize the information-flow scheme for training our network, which explicitly captures the edge information in the node/graph representations and obtains more fine-grained knowledge. Our model achieves an 81.9% ROC-AUC average performance on 661 tasks from seven challenging benchmarks, showing better portability and generalizations. Further visualization studies indicate our model's better representation capacity and provide interpretability.

Recently, pan-sharpening methods based on deep learning (DL) have achieved state-of-the-art results. However, current existing DL-based pan-sharpening methods need to be trained repetitively for different satellite sensors to obtain satisfactory fusion performance and therefore require a large number of training images for each satellite. To deal with these issues, in this paper we propose a unified two-stage spatial and spectral network (UTSN) for pan-sharpening. A branch of networks is constructed for each different satellite, in which the spatial enhancement network (SEN) is shared to improve the spatial details in the fused images from different satellites. A spectral adjustment network (SAN) is employed to capture the spectral characteristics of the specific satellite. Through SAN, the spectral information in the intermediate image from SEN is refined to produce the final fusion results. Such a framework can integrate the datasets from different satellites together for sufficient training of SEN. The proposed method is able to achieve promising pan-sharpening results also for a new satellite with limited training images by only learning a new SAN on the few-shot datasets due to the simple but efficient structure of SAN. The experimental results show that the proposed method can produce state-of-the-art fusion results in both the standard and few-shot cases. The source code is publicly available at https://github.com/RSMagneto/UTSN.

Summary: Gene expression changes over the lifespan and varies among different tissues or cell types. Gene co-expression also changes by sex, age, different tissues or cell types. However, gene expression under the normal state and gene co-expression in the human brain has not been fully defined and quantified. Here we present a database named Brain EXPression Database (BrainEXP) which provides spatiotemporal expression of individual genes and co-expression in normal human brains. BrainEXP consists of 4567 samples from 2863 healthy individuals gathered from existing public databases and our own data, in either microarray or RNA-Seq library types. We mainly provide two analysis results based on the large dataset: (i) basic gene expression across specific brain regions, age ranges and sexes; (ii) co-expression analysis from different platforms. Availability and implementation: http://www.brainexp.org/. Supplementary information: Supplementary data are available at Bioinformatics online.

If we wish to embed assessment for accountability within instruction, we need to better understand the relative contribution of different types of learner data to statistical models that predict scores on assessments used for accountability purposes. The present work scales up and extends predictive models of math test scores from existing literature and specifies six categories of models that incorporate information about student prior knowledge, socio-demographics, and performance within the MATHia intelligent tutoring system. Linear regression and random forest models are learned within each category and generalized over a sample of 23,000+ learners in Grades 6, 7, and 8 over three academic years in Miami-Dade County Public Schools. After briefly exploring hierarchical models of this data, we discuss a variety of technical and practical applications, limitations, and open questions related to this work, especially concerning to the potential use of instructional platforms like MATHia as a replacement for time-consuming standardized tests.