首页期刊导航|中国科学:信息科学(英文版)
期刊信息/Journal information
中国科学:信息科学(英文版)
中国科学:信息科学(英文版)

周光召

月刊

1674-733X

informatics@scichina.org

010-64015683

100717

北京东黄城根北街16号

中国科学:信息科学(英文版)/Journal Science China Information SciencesCSCDCSTPCDEISCI
查看更多>>《中国科学》是中国科学院主办、中国科学杂志社出版的自然科学专业性学术刊物。《中国科学》任务是反映中国自然科学各学科中的最新科研成果,以促进国内外的学术交流。《中国科学》以论文形式报道中国基础研究和应用研究方面具有创造性的、高水平的和有重要意义的科研成果。在国际学术界,《中国科学》作为代表中国最高水平的学术刊物也受到高度重视。国际上最具有权威的检索刊物SCI,多年来一直收录《中国科学》的论文。1999年《中国科学》夺得国家期刊奖的第一名。
正式出版
收录年代

    Super retina TFT based full color microLED display via laser mass transfer

    Xu YANGJinchai LIXuhui PENGChunfeng ZHAO...
    1-15页
    查看更多>>摘要:MicroLED display is considered one of the most promising technologies for next-generation displays.However,the high manufacturing cost has been a major obstacle to its accessibility to the general consumer market,and mass transfer,an essential process to achieve cost-effective manufacturing,has not yet reached commercial maturity.Critical issues,such as microLED chips,transfer equipments,and process materials,need to be addressed for the mass transfer technologies.In this work,we present a 1.63-inch full color microLED display module fabricated with laser mass transfer,which has a pixel density of 403 pixels per inch(PPI),the highest resolution ever achieved in the industry using mass transfer technologies.The laser mass transfer is realized with three process nodes:laser lift-off,laser induced forward transfer,and carrier bonding.Each node has been well explored to improve yields.Insights into the present progress and the future development of the laser mass transfer will be shared in this work.

    Perceptual video quality assessment:a survey

    Xiongkuo MINHuiyu DUANWei SUNYucheng ZHU...
    16-72页
    查看更多>>摘要:Perceptual video quality assessment plays a vital role in the field of video processing due to the existence of quality degradations introduced in various stages of video signal acquisition,compression,transmission and display.With the advancement of Internet communication and cloud service technology,video content and traffic are growing exponentially,which further emphasizes the requirement for accurate and rapid assessment of video quality.Therefore,numerous subjective and objective video quality assess-ment studies have been conducted over the past two decades for both generic videos and specific videos such as streaming,user-generated content,3D,virtual and augmented reality,high dynamic range,high frame rate,audio-visual,etc.This survey provides an up-to-date and comprehensive review of these video quality assessment studies.Specifically,we first review the subjective video quality assessment methodologies and databases,which are necessary for validating the performance of video quality metrics.Second,the objec-tive video quality assessment measures for general purposes are categorized and surveyed according to the methodologies utilized in the quality measures.Third,we overview the objective video quality assessment measures for specific applications and emerging topics.Finally,the performance of the state-of-the-art video quality assessment measures is compared and analyzed.This survey provides a systematic overview of both classical works and recent progress in the realm of video quality assessment,which can help other researchers quickly access the field and conduct relevant research.

    Stochastic normalized gradient descent with momentum for large-batch training

    Shen-Yi ZHAOChang-Wei SHIYin-Peng XIEWu-Jun LI...
    73-87页
    查看更多>>摘要:Stochastic gradient descent(SGD)and its variants have been the dominating optimization meth-ods in machine learning.Compared with SGD with small-batch training,SGD with large-batch training can better utilize the computational power of current multi-core systems such as graphics processing units(GPUs)and can reduce the number of communication rounds in distributed training settings.Thus,SGD with large-batch training has attracted considerable attention.However,existing empirical results showed that large-batch training typically leads to a drop in generalization accuracy.Hence,how to guarantee the gen-eralization ability in large-batch training becomes a challenging task.In this paper,we propose a simple yet effective method,called stochastic normalized gradient descent with momentum(SNGM),for large-batch training.We prove that with the same number of gradient computations,SNGM can adopt a larger batch size than momentum SGD(MSGD),which is one of the most widely used variants of SGD,to converge to an e-stationary point.Empirical results on deep learning verify that when adopting the same large batch size,SNGM can achieve better test accuracy than MSGD and other state-of-the-art large-batch training methods.

    Residual diverse ensemble for long-tailed multi-label text classification

    Jiangxin SHITong WEIYufeng LI
    88-101页
    查看更多>>摘要:Long-tailed multi-label text classification aims to identify a subset of relevant labels from a large candidate label set,where the training datasets usually follow long-tailed label distributions.Many of the previous studies have treated head and tail labels equally,resulting in unsatisfactory performance for identifying tail labels.To address this issue,this paper proposes a novel learning method that combines arbitrary models with two steps.The first step is the"diverse ensemble"that encourages diverse predictions among multiple shallow classifiers,particularly on tail labels,and can improve the generalization of tail labels.The second is the"error correction"that takes advantage of accurate predictions on head labels by the base model and approximates its residual errors for tail labels.Thus,it enables the"diverse ensemble"to focus on optimizing the tail label performance.This overall procedure is called residual diverse ensemble(RDE).RDE is implemented via a single-hidden-layer perceptron and can be used for scaling up to hundreds of thousands of labels.We empirically show that RDE consistently improves many existing models with considerable performance gains on benchmark datasets,especially with respect to the propensity-scored evaluation metrics.Moreover,RDE converges in less than 30 training epochs without increasing the computational overhead.

    RCCA-SM9:securing SM9 on corrupted machines

    Rongmao CHENJinrong CHENXinyi HUANGYi WANG...
    102-116页
    查看更多>>摘要:The SM9 identity-based encryption(IBE)scheme is a cryptographic standard used in China,and has been incorporated into the ISO/IEC standard in 2021.This work primarily proposes a countermeasure to secure the SM9 IBE scheme if its implementation is tampered with or deviated from the standard specification.Such attacks,known as subversion attacks,are feasible and powerful in real-world cryptographic application scenarios.Our goal is to design a subversion-resilient variant of the SM9 IBE scheme,primarily using the cryptographic reverse firewall(CRF)proposed by Mironov and Stephens-Davidowitz at EUROCRYPT 2015.A CRF can sanitize cryptographic transcripts to eliminate covert channels,necessitating that the underlying primitive be rerandomizable.Unfortunately,the rerandomizability of the SM9 IBE scheme is disabled for ensuring security against chosen ciphertext attack(CCA).Hence,we shift our focus to a relaxed version of CCA security called RCCA security,offering security guarantees comparable to CCA security while allowing for ciphertext rerandomization.For this purpose,we design an efficient and RCCA-secure variant of the SM9 IBE scheme with provable security that can integrate with CRFs to achieve subversion resilience.

    PointSmile:point self-supervised learning via curriculum mutual information

    Xin LIMingqiang WEISongcan CHEN
    117-131页
    查看更多>>摘要:Self-supervised learning is attracting significant attention from researchers in the point cloud processing field.However,due to the natural sparsity and irregularity of point clouds,effectively extract-ing discriminative and transferable features for efficient training on downstream tasks remains an unsolved challenge.Consequently,we propose PointSmile,a reconstruction-free self-supervised learning paradigm by maximizing curriculum mutual information(CMI)across the replicas of point cloud objects.From the perspective of how-and-what-to-learn,PointSmile is designed to imitate human curriculum learning,i.e.,starting with easier topics in a curriculum and gradually progressing to learning more complex topics in the curriculum.To solve"how-to-learn",we introduce curriculum data augmentation(CDA)of point clouds.CDA encourages PointSmile to follow a learning path that starts from learning easy data samples and pro-gresses to learning hard data samples,such that the latent space can be dynamically affected to create better embeddings.To solve"what-to-learn",we propose maximizing both feature-and class-wise CMI to better extract discriminative features of point clouds.Unlike most existing methods,PointSmile does not require a pretext task or cross-modal data to yield rich latent representations;additionally,it can be easily transferred to various backbones.We demonstrate the effectiveness and robustness of PointSmile in downstream tasks such as object classification and segmentation.The study results show that PointSmile outperforms existing self-supervised methods and compares favorably with popular fully supervised methods on various standard architectures.The code is available at https://github.com/theaalee/PointSmile.

    TPpred-SC:multi-functional therapeutic peptide prediction based on multi-label supervised contrastive learning

    Ke YANHongwu LVJiangyi SHAOShutao CHEN...
    132-143页
    查看更多>>摘要:Therapeutic peptides contribute significantly to human health and have the potential for per-sonalized medicine.The prediction for the therapeutic peptides is beneficial and emerging for the discovery of drugs.Although several computational approaches have emerged to discern the functions of therapeu-tic peptides,predicting multi-functional therapeutic peptide types is challenging.In this research,a novel approach termed TPpred-SC has been introduced.This method leverages a pretrained protein language model alongside multi-label supervised contrastive learning to predict multi-functional therapeutic peptides.The framework incorporates sequential semantic information directly from large-scale protein sequences in TAPE.Then,TPpred-SC exploits multi-label supervised contrastive learning to enhance the representation of peptide sequences for imbalanced multi-label therapeutic peptide prediction.The experimental findings demonstrate that TPpred-SC achieves superior performance compared to existing related methods.To serve our work more efficiently,the web server of TPpred-SC can be accessed at http://bliulab.net/TPpred-SC.

    SBSM-Pro:support bio-sequence machine for proteins

    Yizheng WANGYixiao ZHAIYijie DINGQuan ZOU...
    144-159页
    查看更多>>摘要:Proteins play a pivotal role in biological systems.The use of machine learning algorithms for protein classification can assist and even guide biological experiments,offering crucial insights for biotech-nological applications.We introduce the support bio-sequence machine for proteins(SBSM-Pro),a model purpose-built for the classification of biological sequences.This model starts with raw sequences and groups amino acids based on their physicochemical properties.It incorporates sequence alignment to measure the similarities between proteins and uses a novel multiple kernel learning(MKL)approach to integrate various types of information,utilizing support vector machines for classification prediction.The results indicate that our model demonstrates commendable performance across ten datasets in terms of the identification of protein function and post translational modification.This research not only exemplifies state-of-the-art work in protein classification but also paves avenues for new directions in this domain,representing a beneficial endeavor in the development of platforms tailored for the classification of biological sequences.SBSM-Pro is available for access at http://lab.malab.cn/soft/SBSM-Pro/.

    Improved dynamic regret of distributed online multiple Frank-Wolfe convex optimization

    Wentao ZHANGYang SHIBaoyong ZHANGDeming YUAN...
    160-175页
    查看更多>>摘要:In this paper,we explore a distributed online convex optimization problem over a time-varying multi-agent network.The network aims to minimize a global loss function through local computation and communication with neighboring agents.To effectively handle the optimization problem which involves high-dimensional and structural constraint sets,we develop a distributed online multiple Frank-Wolfe algorithm that circumvents the expensive computational cost associated with projection operations.The dynamic regret bounds are established as O(T1-γ+HT)with the linear oracle number O(T1+γ),which depends on the horizon(total iteration number)T,the function variation HT,and the tuning parameter 0<γ<1.In particular,when the prior knowledge of HT and T is available,the bound can be enhanced to O(1+HT).Moreover,we explore the significant advantages provided by the multiple iteration technique and reveal a trade-off between dynamic regret bound,computational cost,and communication cost.Finally,the performance of our algorithm is validated and compared through the distributed online ridge regression problems with two constraint sets.

    Coalition formation problem:a capability-centric analysis and general model

    Jie CHENMiao GUOBin XINQing WANG...
    176-189页
    查看更多>>摘要:Coalition formation(CF)refers to reasonably organizing robots and/or humans to form coalitions that can satisfy mission requirements,attracting more and more attention in many fields such as multi-robot collaboration and human-robot collaboration.However,the analysis on CF problems remains limited.To provide a valuable study reference for researchers interested in CF,this paper proposed a capability-centric analysis of the CF problem.The key problem elements of CF are firstly extracted by referencing the concepts of the 5W1H method.That is,objects(who)form coalitions(what)to accomplish missions(why)by aggregating capabilities(how)in a specific environment(where-when).Then,a multi-view analysis of these elements and their correlation in terms of capabilities is proposed through various logic diagrams,structure charts,etc.Finally,to facilitate a deeper understanding of capability-centric CF,a general mathematical model is constructed,demonstrating how the different concepts discussed in this analysis contribute to the overall model.