Hefei University of Technology

State-of-the-art object detection networks depend on region proposal\nalgorithms to hypothesize object locations. Advances like SPPnet and Fast R-CNN\nhave reduced the running time of these detection networks, exposing region\nproposal computation as a bottleneck. In this work, we introduce a Region\nProposal Network (RPN) that shares full-image convolutional features with the\ndetection network, thus enabling nearly cost-free region proposals. An RPN is a\nfully convolutional network that simultaneously predicts object bounds and\nobjectness scores at each position. The RPN is trained end-to-end to generate\nhigh-quality region proposals, which are used by Fast R-CNN for detection. We\nfurther merge RPN and Fast R-CNN into a single network by sharing their\nconvolutional features---using the recently popular terminology of neural\nnetworks with 'attention' mechanisms, the RPN component tells the unified\nnetwork where to look. For the very deep VGG-16 model, our detection system has\na frame rate of 5fps (including all steps) on a GPU, while achieving\nstate-of-the-art object detection accuracy on PASCAL VOC 2007, 2012, and MS\nCOCO datasets with only 300 proposals per image. In ILSVRC and COCO 2015\ncompetitions, Faster R-CNN and RPN are the foundations of the 1st-place winning\nentries in several tracks. Code has been made publicly available.\n

Due to object detection's close relationship with video analysis and image understanding, it has attracted much research attention in recent years. Traditional object detection methods are built on handcrafted features and shallow trainable architectures. Their performance easily stagnates by constructing complex ensembles that combine multiple low-level image features with high-level context from object detectors and scene classifiers. With the rapid development in deep learning, more powerful tools, which are able to learn semantic, high-level, deeper features, are introduced to address the problems existing in traditional architectures. These models behave differently in network architecture, training strategy, and optimization function. In this paper, we provide a review of deep learning-based object detection frameworks. Our review begins with a brief introduction on the history of deep learning and its representative tool, namely, the convolutional neural network. Then, we focus on typical generic object detection architectures along with some modifications and useful tricks to improve detection performance further. As distinct specific detection tasks exhibit different characteristics, we also briefly survey several specific tasks, including salient object detection, face detection, and pedestrian detection. Experimental analyses are also provided to compare various methods and draw some meaningful conclusions. Finally, several promising directions and tasks are provided to serve as guidelines for future work in both object detection and relevant neural network-based learning systems.

Graph Convolution Network (GCN) has become new state-of-the-art for collaborative filtering. Nevertheless, the reasons of its effectiveness for recommendation are not well understood. Existing work that adapts GCN to recommendation lacks thorough ablation analyses on GCN, which is originally designed for graph classification tasks and equipped with many neural network operations. However, we empirically find that the two most common designs in GCNs -- feature transformation and nonlinear activation -- contribute little to the performance of collaborative filtering. Even worse, including them adds to the difficulty of training and degrades recommendation performance.

Abstract Somatic mutations in cancer genomes are caused by multiple mutational processes, each of which generates a characteristic mutational signature 1 . Here, as part of the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium 2 of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA), we characterized mutational signatures using 84,729,690 somatic mutations from 4,645 whole-genome and 19,184 exome sequences that encompass most types of cancer. We identified 49 single-base-substitution, 11 doublet-base-substitution, 4 clustered-base-substitution and 17 small insertion-and-deletion signatures. The substantial size of our dataset, compared with previous analyses 3–15 , enabled the discovery of new signatures, the separation of overlapping signatures and the decomposition of signatures into components that may represent associated—but distinct—DNA damage, repair and/or replication mechanisms. By estimating the contribution of each signature to the mutational catalogues of individual cancer genomes, we revealed associations of signatures to exogenous or endogenous exposures, as well as to defective DNA-maintenance processes. However, many signatures are of unknown cause. This analysis provides a systematic perspective on the repertoire of mutational processes that contribute to the development of human cancer.

Abstract Cancer is driven by genetic change, and the advent of massively parallel sequencing has enabled systematic documentation of this variation at the whole-genome scale 1–3 . Here we report the integrative analysis of 2,658 whole-cancer genomes and their matching normal tissues across 38 tumour types from the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA). We describe the generation of the PCAWG resource, facilitated by international data sharing using compute clouds. On average, cancer genomes contained 4–5 driver mutations when combining coding and non-coding genomic elements; however, in around 5% of cases no drivers were identified, suggesting that cancer driver discovery is not yet complete. Chromothripsis, in which many clustered structural variants arise in a single catastrophic event, is frequently an early event in tumour evolution; in acral melanoma, for example, these events precede most somatic point mutations and affect several cancer-associated genes simultaneously. Cancers with abnormal telomere maintenance often originate from tissues with low replicative activity and show several mechanisms of preventing telomere attrition to critical levels. Common and rare germline variants affect patterns of somatic mutation, including point mutations, structural variants and somatic retrotransposition. A collection of papers from the PCAWG Consortium describes non-coding mutations that drive cancer beyond those in the TERT promoter 4 ; identifies new signatures of mutational processes that cause base substitutions, small insertions and deletions and structural variation 5,6 ; analyses timings and patterns of tumour evolution 7 ; describes the diverse transcriptional consequences of somatic mutation on splicing, expression levels, fusion genes and promoter activity 8,9 ; and evaluates a range of more-specialized features of cancer genomes 8,10–18 .

Learning vector representations (aka. embeddings) of users and items lies at the core of modern recommender systems. Ranging from early matrix factorization to recently emerged deep learning based methods, existing efforts typically obtain a user's (or an item's) embedding by mapping from pre-existing features that describe the user (or the item), such as ID and attributes. We argue that an inherent drawback of such methods is that, the collaborative signal, which is latent in user-item interactions, is not encoded in the embedding process. As such, the resultant embeddings may not be sufficient to capture the collaborative filtering effect.

Big Data concern large-volume, complex, growing data sets with multiple, autonomous sources. With the fast development of networking, data storage, and the data collection capacity, Big Data are now rapidly expanding in all science and engineering domains, including physical, biological and biomedical sciences. This paper presents a HACE theorem that characterizes the features of the Big Data revolution, and proposes a Big Data processing model, from the data mining perspective. This data-driven model involves demand-driven aggregation of information sources, mining and analysis, user interest modeling, and security and privacy considerations. We analyze the challenging issues in the data-driven model and also in the Big Data revolution.

Deep learning (DL) algorithms have seen a massive rise in popularity for remote-sensing image analysis over the past few years. In this study, the major DL concepts pertinent to remote-sensing are introduced, and more than 200 publications in this field, most of which were published during the last two years, are reviewed and analyzed. Initially, a meta-analysis was conducted to analyze the status of remote sensing DL studies in terms of the study targets, DL model(s) used, image spatial resolution(s), type of study area, and level of classification accuracy achieved. Subsequently, a detailed review is conducted to describe/discuss how DL has been applied for remote sensing image analysis tasks including image fusion, image registration, scene classification, object detection, land use and land cover (LULC) classification, segmentation, and object-based image analysis (OBIA). This review covers nearly every application and technology in the field of remote sensing, ranging from preprocessing to mapping. Finally, a conclusion regarding the current state-of-the art methods, a critical conclusion on open challenges, and directions for future research are presented.

Because of their high theoretical energy density and low cost, lithium–sulfur (Li–S) batteries are promising next-generation energy storage devices. The electrochemical performance of Li–S batteries largely depends on the efficient reversible conversion of Li polysulfides to Li2S in discharge and to elemental S during charging. Here, we report on our discovery that monodisperse cobalt atoms embedded in nitrogen-doped graphene (Co–N/G) can trigger the surface-mediated reaction of Li polysulfides. Using a combination of operando X-ray absorption spectroscopy and first-principles calculation, we reveal that the Co–N–C coordination center serves as a bifunctional electrocatalyst to facilitate both the formation and the decomposition of Li2S in discharge and charge processes, respectively. The S@Co–N/G composite, with a high S mass ratio of 90 wt %, can deliver a gravimetric capacity of 1210 mAh g–1, and it exhibits an areal capacity of 5.1 mAh cm–2 with capacity fading rate of 0.029% per cycle over 100 cycles at 0.2 C at S loading of 6.0 mg cm–2 on the electrode disk.

Abstract Two-dimensional (2D) materials have emerged as promising candidates for various optoelectronic applications based on their diverse electronic properties, ranging from insulating to superconducting. However, cooperative phenomena such as ferroelectricity in the 2D limit have not been well explored. Here, we report room-temperature ferroelectricity in 2D CuInP 2 S 6 (CIPS) with a transition temperature of ∼320 K. Switchable polarization is observed in thin CIPS of ∼4 nm. To demonstrate the potential of this 2D ferroelectric material, we prepare a van der Waals (vdW) ferroelectric diode formed by CIPS/Si heterostructure, which shows good memory behaviour with on/off ratio of ∼100. The addition of ferroelectricity to the 2D family opens up possibilities for numerous novel applications, including sensors, actuators, non-volatile memory devices, and various vdW heterostructures based on 2D ferroelectricity.

Abstract Cancer develops through a process of somatic evolution 1,2 . Sequencing data from a single biopsy represent a snapshot of this process that can reveal the timing of specific genomic aberrations and the changing influence of mutational processes 3 . Here, by whole-genome sequencing analysis of 2,658 cancers as part of the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA) 4 , we reconstruct the life history and evolution of mutational processes and driver mutation sequences of 38 types of cancer. Early oncogenesis is characterized by mutations in a constrained set of driver genes, and specific copy number gains, such as trisomy 7 in glioblastoma and isochromosome 17q in medulloblastoma. The mutational spectrum changes significantly throughout tumour evolution in 40% of samples. A nearly fourfold diversification of driver genes and increased genomic instability are features of later stages. Copy number alterations often occur in mitotic crises, and lead to simultaneous gains of chromosomal segments. Timing analyses suggest that driver mutations often precede diagnosis by many years, if not decades. Together, these results determine the evolutionary trajectories of cancer, and highlight opportunities for early cancer detection.

Free-standing graphene paper with a grey metallic luster has been fabricated for the first time, by a convenient one-step method on a large scale. Herein, the assembly of graphene oxide dispersion into ordered paper occurs simultaneously with the chemical reduction of graphene oxide to graphene. The graphene paper presents the advantages of good flexibility, low weight (0.2 g cm−3) and high electrical conductivity (15 Ω sq−1). Moreover, the size and shape of the graphene paper are freely exchanged for those of the Teflon substrate used. The flexible graphene–PANI paper subsequently exhibits excellent supercapacitor performance with an enhanced specific capacitance (763 F g−1) and good cycling stability by electropolymerization of PANI nanorods on the above graphene paper. The method presented here shows great promise for the development of low-cost electrode materials in potential energy storage devices.

This paper proposes a double-sided LCC compensation network and its tuning method for wireless power transfer (WPT). With the proposed topology and its tuning method, the resonant frequency is irrelevant with the coupling coefficient between the two coils and is also independent of the load condition, which means that the system can work at a constant switching frequency. Analysis in frequency domain is given to show the characteristics of the proposed method. We also propose a method to tune the network to realize zero voltage switching (ZVS) for the Primary-side switches. Simulation and experimental results verified analysis and validity of the proposed compensation network and the tuning method. A wireless charging system with output power of up to 7.7 kW for electric vehicles was built, and 96% efficiency from dc power source to battery load is achieved.

Abstract A key mutational process in cancer is structural variation, in which rearrangements delete, amplify or reorder genomic segments that range in size from kilobases to whole chromosomes 1–7 . Here we develop methods to group, classify and describe somatic structural variants, using data from the Pan-Cancer Analysis of Whole Genomes (PCAWG) Consortium of the International Cancer Genome Consortium (ICGC) and The Cancer Genome Atlas (TCGA), which aggregated whole-genome sequencing data from 2,658 cancers across 38 tumour types 8 . Sixteen signatures of structural variation emerged. Deletions have a multimodal size distribution, assort unevenly across tumour types and patients, are enriched in late-replicating regions and correlate with inversions. Tandem duplications also have a multimodal size distribution, but are enriched in early-replicating regions—as are unbalanced translocations. Replication-based mechanisms of rearrangement generate varied chromosomal structures with low-level copy-number gains and frequent inverted rearrangements. One prominent structure consists of 2–7 templates copied from distinct regions of the genome strung together within one locus. Such cycles of templated insertions correlate with tandem duplications, and—in liver cancer—frequently activate the telomerase gene TERT . A wide variety of rearrangement processes are active in cancer, which generate complex configurations of the genome upon which selection can act.

As a popular signal modeling technique, sparse representation (SR) has achieved great success in image fusion over the last few years with a number of effective algorithms being proposed. However, due to the patch-based manner applied in sparse coding, most existing SR-based fusion methods suffer from two drawbacks, namely, limited ability in detail preservation and high sensitivity to misregistration, while these two issues are of great concern in image fusion. In this letter, we introduce a recently emerged signal decomposition model known as convolutional sparse representation (CSR) into image fusion to address this problem, which is motivated by the observation that the CSR model can effectively overcome the above two drawbacks. We propose a CSR-based image fusion framework, in which each source image is decomposed into a base layer and a detail layer, for multifocus image fusion and multimodal image fusion. Experimental results demonstrate that the proposed fusion methods clearly outperform the SR-based methods in terms of both objective assessment and visual quality.

Large language models (LLMs), such as ChatGPT and GPT4, are making new waves in the field of natural language processing and artificial intelligence, due to their emergent ability and generalizability. However, LLMs are black-box models, which often fall short of capturing and accessing factual knowledge. In contrast, Knowledge Graphs (KGs), Wikipedia and Huapu for example, are structured knowledge models that explicitly store rich factual knowledge. KGs can enhance LLMs by providing external knowledge for inference and interpretability. Meanwhile, KGs are difficult to construct and evolve by nature, which challenges the existing methods in KGs to generate new facts and represent unseen knowledge. Therefore, it is complementary to unify LLMs and KGs together and simultaneously leverage their advantages. In this article, we present a forward-looking roadmap for the unification of LLMs and KGs. Our roadmap consists of three general frameworks, namely, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1) KG-enhanced LLMs,</i> which incorporate KGs during the pre-training and inference phases of LLMs, or for the purpose of enhancing understanding of the knowledge learned by LLMs; <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2) LLM-augmented KGs,</i> that leverage LLMs for different KG tasks such as embedding, completion, construction, graph-to-text generation, and question answering; and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3) Synergized LLMs + KGs</i> , in which LLMs and KGs play equal roles and work in a mutually beneficial way to enhance both LLMs and KGs for bidirectional reasoning driven by both data and knowledge. We review and summarize existing efforts within these three frameworks in our roadmap and pinpoint their future research directions.

Sodium manganese hexacyanoferrates (NMHFCs) synthesized in aqueous solution at room temperature show high reversible capacity and outstanding rate capability as cathodes for a rechargeable sodium-ion battery (SIB). Earth-abundant elements and a low-cost synthesis route make these NMHFCs promising cathodes for SIBs, independent of natural lithium sources.

Lactic acid bacteria (LAB) constitute a heterogeneous group of microorganisms that produce lactic acid as the major product during the fermentation process. LAB are Gram-positive bacteria with great biotechnological potential in the food industry. They can produce bacteriocins, which are proteinaceous antimicrobial molecules with a diverse genetic origin, posttranslationally modified or not, that can help the producer organism to outcompete other bacterial species. In this review, we focus on the various types of bacteriocins that can be found in LAB and the organization and regulation of the gene clusters responsible for their production and biosynthesis, and consider the food applications of the prototype bacteriocins from LAB. Furthermore, we propose a revised classification of bacteriocins that can accommodate the increasing number of classes reported over the last years.

A deep UV light photodetector is assembled by coating multilayer graphene on beta-gallium oxide (β-Ga2O3) wafer. Optoelectronic analysis reveals that the heterojunction device is virtually blind to light illumination with wavelength longer than 280 nm, but is highly sensitive to 254 nm light with very good stability and reproducibility. As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Abstract Sodium‐ion batteries (SIBs) have a promising application prospect for energy storage systems due to the abundant resource. Amorphous carbon with high electronic conductivity and high surface area is likely to be the most promising anode material for SIBs. However, the rate capability of amorphous carbon in SIBs is still a big challenge because of the sluggish kinetics of Na + ions. Herein, a three‐dimensional amorphous carbon (3DAC) with controlled porous and disordered structures is synthesized via a facile NaCl template‐assisted method. Combination of open porous structures of 3DAC, the increased disordered structures can not only facilitate the diffusion of Na + ions but also enhance the reversible capacity of Na storage. When applied as anode materials for SIBs, 3DAC exhibits excellent rate capability (66 mA h g −1 at 9.6 A g −1 ) and high reversible capacity (280 mA h g −1 at a low current density of 0.03 A g −1 ). Moreover, the controlled porous structures by the NaCl template method provide an appropriate specific surface area, which contributes to a relatively high initial Coulombic efficiency of 75%. Additionally, the high‐rate 3DAC material is prepared via a green approach originating from low‐cost pitch and NaCl template, demonstrating an appealing development of carbon anode materials for SIBs.