Search

Semi-supervised learning has achieved extraordinary success in prevalent image-classification benchmarks. However, a class-balanced distribution that differs notably from real-world data distribution is required. In general, models trained under class-imbalanced semi-supervised learning conditions are severely biased towards the majority classes. To address this issue, we propose a novel framework called ABAE by implanting an Auxiliary Balanced AutoEncoder branch into existing semi-supervised learning algorithms. Considering that adaptive feature augmentation for different classes can alleviate confirmation bias, we devise a class-aware reconstruction loss to train the AutoEncoder module. To further smooth the output, we adopt a graph-based label propagation scheme at the end of the AutoEncoder. Extensive experiments on CIFAR-10/100-LT, SVHN-LT and Small ImageNet-127 demonstrate the effectiveness of ABAE. • Propose an Auxiliary Balanced AutoEncoder branch(ABAE) with a class-aware reconstruction loss for CISSL. • Utilize the adaptive feature augmentation for different classes to alleviate the confirmation bias. • Adopt a graph-based label propagation scheme to smooth the output of the ABAE. • Demonstrate the effectiveness of the ABAE through experiments on various imbalanced datasets. [ABSTRACT FROM AUTHOR]

An infrared photodetector is a critical component that detects, identifies, and tracks complex targets in a detection system. Infrared photodetectors based on 3D bulk materials are widely applied in national defense, military, communications, and astronomy fields. The complex application environment requires higher performance and multi-dimensional capability. The emergence of 2D materials has brought new possibilities to develop next-generation infrared detectors. However, the inherent thickness limitations and the immature preparation of 2D materials still lead to low quantum efficiency and slow response speeds. This review summarizes 2D/3D hybrid van der Waals heterojunctions for infrared photodetection. First, the physical properties of 2D and 3D materials related to detection capability, including thickness, band gap, absorption band, quantum efficiency, and carrier mobility, are summarized. Then, the primary research progress of 2D/3D infrared detectors is reviewed from performance improvement (broadband, high-responsivity, fast response) and new functional devices (two-color detectors, polarization detectors). Importantly, combining low-doped 3D and flexible 2D materials can effectively improve the responsivity and detection speed due to a significant depletion region width. Furthermore, combining the anisotropic 2D lattice structure and high absorbance of 3D materials provides a new strategy in high-performance polarization detectors. This paper offers prospects for developing 2D/3D high-performance infrared detection technology. [ABSTRACT FROM AUTHOR]

Background: Soil salinization is a major abiotic environmental stress factor threatening crop production throughout the world. Salt stress drastically affects the growth, development, and grain yield of rice (Oryza sativa L.), and the improvement of rice tolerance to salt stress is a desirable approach for meeting increasing food demand. Receptor-like cytoplasmic kinases (RLCKs) play essential roles in plant growth, development and responses to environmental stresses. However, little is known about their functions in salt stress. Previous reports have demonstrated that overexpression of an RLCK gene SALT TOLERANCE KINASE (STK) enhances salt tolerance in rice, and that STK may regulate the expression of GST (Glutathione S-transferase) genes. Results: The expression of STK was rapidly induced by ABA. STK was highest expressed in the stem at the heading stage. STK was localized at the plasma membrane. Overexpression of STK in rice increased tolerance to salt stress and oxidative stress by increasing ROS scavenging ability and ABA sensitivity. In contrast, CRISPR/Cas9-mediated knockout of STK increased the sensitivity of rice to salt stress and oxidative stress. Transcriptome sequencing analysis suggested that STK increased the expression of GST genes (LOC_Os03g17480, LOC_Os10g38140 and LOC_Os10g38710) under salt stress. Reverse transcription quantitative PCR (RT-qPCR) suggested that four stress-related genes may be regulated by STK including OsABAR1, Os3BGlu6, OSBZ8 and OsSIK1. Conclusions: These findings suggest that STK plays a positive regulatory role in salt stress tolerance by inducing antioxidant defense and associated with the ABA signaling pathway in rice. [ABSTRACT FROM AUTHOR]

Summary: Rice blast and bacterial blight represent two of major diseases having devastating impact on the yield of rice in most rice‐growing countries. Developments of resistant cultivars are the most economic and effective strategy to control these diseases. Here, we used CRISPR/Cas9‐mediated gene editing to rapidly install mutations in three known broad‐spectrum blast‐resistant genes, Bsr‐d1, Pi21 and ERF922, in an indica thermosensitive genic male sterile (TGMS) rice line Longke638S (LK638S). We obtained transgene‐free homozygous single or triple mutants in T1 generations. While all single and triple mutants showed increased resistance to rice blast compared with wild type, the erf922 mutants displayed the strongest blast resistance similar with triple mutants. Surprisingly, we found that Pi21 or ERF922 single mutants conferred enhanced resistance to most of tested bacterial blight. Both resistances in mutants were attribute to the up‐regulation of SA‐ and JA‐pathway associated genes. Moreover, phenotypic analysis of these single mutants in paddy fields revealed that there were no trade‐offs between resistances and main agricultural traits. Together, our study provides a rapid and effective way to generate rice varieties with resistance to both rice blast and bacterial blight. [ABSTRACT FROM AUTHOR]

Purpose: Impairments in intra- and inter-hemispheric information transfer circuits have been reported in patients with major depressive disorder (MDD). However, the specific anomalous connection (intra- and/or inter-hemispheric) and hemisphere (left and/or right) in which this connection plays a more dominant role in the pathogenic mechanism underlying MDD are still poorly understood. Patients and Methods: Structural magnetic resonance imaging and resting-state functional magnetic resonance imaging were performed in 33 patients with MDD and 33 healthy controls. The intra- and inter-hemispheric functional connectivity (FC) strength in the default mode network areas and volume of the callosal subregions were computed using independent samples t-tests. The partial correlations between the volumes and FCs were also computed. Results: The patients with MDD had smaller volumes in the genu of the corpus callosum than the controls. The intrahemispheric FCs of the bilateral posterior cingulate gyrus, left precuneus, left medial superior frontal gyrus, left medial orbitofrontal gyrus, left angular gyrus and left middle temporal gyrus, and interhemispheric FCs of the bilateral posterior cingulate gyrus in the patients with MDD were lower than those in the controls. Moreover, the intrahemispheric FCs of the precuneus and interhemispheric FCs of middle frontal gyrus, orbital middle frontal gyrus, and anterior cingulate gyrus in the patients with MDD showed right-lateralized asymmetry, which were opposite from the asymmetry patterns observed in the controls. The functional asymmetry of the anterior cingulate gyrus was correlated with the volume of the genu of the corpus callosum and disease duration. Conclusion: These findings provide robust evidence that intra- and inter-hemispheric disconnections are involved in MDD, and that functional disruptions in the left hemisphere may be more relevant to the pathophysiology of MDD. Furthermore, imbalanced interhemispheric exchanges may contribute to the anatomical deficits in the corpus callosum in patients with MDD. [ABSTRACT FROM AUTHOR]

We report a detailed analysis of neutron-induced multibit-upset (MBU) clusters measured from flip-flop arrays implemented in a 14-nm trigate CMOS. Depending on the strike location, charge collection efficiency, and circuit topology, the MBU clusters are characterized in terms of size and span, and a qualitative first-order analysis has been presented. A novel MBU analysis framework has been demonstrated that uses a weighted sliding window to characterize MBU clusters efficiently and accurately with minimal double-counting or mischaracterization of cluster size. To further explain the relative MBU cross sections, the unique MBU patterns extracted from measured data have been studied and analyzed to find layout dependencies. The results show the strong correlation between the internode proximity and MBUs. The analysis shows a higher soft error rate (SER) cross section for smaller MBU cluster size and smaller span while the bigger clusters have lower contribution toward overall MBU SER. [ABSTRACT FROM AUTHOR]

Circuit reliability issues such as bias temperature instability, hot carrier injection, time-dependent dielectric breakdown, electromigration, and random telegraph noise have become a growing concern with technology scaling. Precise measurements of circuit degradation induced by these reliability mechanisms are a key aspect of robust design. This article reviews several unique test-chip designs that demonstrate the benefits of utilizing on-chip logic and a simple test interface to automate circuit aging experiments. This new class of compact on-chip sensors can reveal important aspects of circuit aging that would otherwise be impossible to measure, facilitate the collection of reliability data from systems deployed in the field, and eventually lead us down the path to real-time aging compensation in future processors. [ABSTRACT FROM PUBLISHER]