BViT: Broad Attention-Based Vision Transformer.

Recent works have demonstrated that transformer can achieve promising performance in computer vision, by exploiting the relationship among image patches with self-attention. They only consider the attention in a single feature layer, but ignore the complementarity of attention in different layers. In this article, we propose broad attention to improve the performance by incorporating the attention relationship of different layers for vision transformer (ViT), which is called BViT. The broad attention is implemented by broad connection and parameter-free attention. Broad connection of each transformer layer promotes the transmission and integration of information for BViT. Without introducing additional trainable parameters, parameter-free attention jointly focuses on the already available attention information in different layers for extracting useful information and building their relationship. Experiments on image classification tasks demonstrate that BViT delivers superior accuracy of 75.0%/81.6% top-1 accuracy on ImageNet with 5M/22M parameters. Moreover, we transfer BViT to downstream object recognition benchmarks to achieve 98.9% and 89.9% on CIFAR10 and CIFAR100, respectively, that exceed ViT with fewer parameters. For the generalization test, the broad attention in Swin Transformer, T2T-ViT and LVT also brings an improvement of more than 1%. To sum up, broad attention is promising to promote the performance of attention-based models. Code and pretrained models are available at https://github.com/DRL/BViT.

BiFNet: Bidirectional Fusion Network for Road Segmentation.

Multisensor fusion-based road segmentation plays an important role in the intelligent driving system since it provides a drivable area. The existing mainstream fusion method is mainly to feature fusion in the image space domain which causes the perspective compression of the road and damages the performance of the distant road. Considering the bird's eye views (BEVs) of the LiDAR remains the space structure in the horizontal plane, this article proposes a bidirectional fusion network (BiFNet) to fuse the image and BEV of the point cloud. The network consists of two modules: 1) the dense space transformation (DST) module, which solves the mutual conversion between the camera image space and BEV space and 2) the context-based feature fusion module, which fuses the different sensors information based on the scenes from corresponding features. This method has achieved competitive results on the KITTI dataset.

ModuleNet: Knowledge-Inherited Neural Architecture Search.

Although neural the architecture search (NAS) can bring improvement to deep models, it always neglects precious knowledge of existing models. The computation and time costing property in NAS also means that we should not start from scratch to search, but make every attempt to reuse the existing knowledge. In this article, we discuss what kind of knowledge in a model can and should be used for a new architecture design. Then, we propose a new NAS algorithm, namely, ModuleNet, which can fully inherit knowledge from the existing convolutional neural networks. To make full use of the existing models, we decompose existing models into different modules, which also keep their weights, consisting of a knowledge base. Then, we sample and search for a new architecture according to the knowledge base. Unlike previous search algorithms, and benefiting from inherited knowledge, our method is able to directly search for architectures in the macrospace by the NSGA-II algorithm without tuning parameters in these modules. Experiments show that our strategy can efficiently evaluate the performance of a new architecture even without tuning weights in convolutional layers. With the help of knowledge we inherited, our search results can always achieve better performance on various datasets (CIFAR10, CIFAR100, and ImageNet) over original architectures.

BNAS: Efficient Neural Architecture Search Using Broad Scalable Architecture.

Efficient neural architecture search (ENAS) achieves novel efficiency for learning architecture with high-performance via parameter sharing and reinforcement learning (RL). In the phase of architecture search, ENAS employs deep scalable architecture as search space whose training process consumes most of the search cost. Moreover, time-consuming model training is proportional to the depth of deep scalable architecture. Through experiments using ENAS on CIFAR-10, we find that layer reduction of scalable architecture is an effective way to accelerate the search process of ENAS but suffers from a prohibitive performance drop in the phase of architecture estimation. In this article, we propose a broad neural architecture search (BNAS) where we elaborately design broad scalable architecture dubbed broad convolutional neural network (BCNN) to solve the above issue. On the one hand, the proposed broad scalable architecture has fast training speed due to its shallow topology. Moreover, we also adopt RL and parameter sharing used in ENAS as the optimization strategy of BNAS. Hence, the proposed approach can achieve higher search efficiency. On the other hand, the broad scalable architecture extracts multi-scale features and enhancement representations, and feeds them into global average pooling (GAP) layer to yield more reasonable and comprehensive representations. Therefore, the performance of broad scalable architecture can be promised. In particular, we also develop two variants for BNAS that modify the topology of BCNN. In order to verify the effectiveness of BNAS, several experiments are performed and experimental results show that 1) BNAS delivers 0.19 days which is 2.37× less expensive than ENAS who ranks the best in RL-based NAS approaches; 2) compared with small-size (0.5 million parameters) and medium-size (1.1 million parameters) models, the architecture learned by BNAS obtains state-of-the-art performance (3.58% and 3.24% test error) on CIFAR-10; and 3) the learned architecture achieves 25.3% top-1 error on ImageNet just using 3.9 million parameters.

Boost 3-D Object Detection via Point Clouds Segmentation and Fused 3-D GIoU-L1 Loss.

The 3-D object detection is crucial for many real-world applications, attracting many researchers' attention. Beyond 2-D object detection, 3-D object detection usually needs to extract appearance, depth, position, and orientation information from light detection and ranging (LiDAR) and camera sensors. However, due to more degrees of freedom and vertices, existing detection methods that directly transform from 2-D to 3-D still face several challenges, such as exploding increase of anchors' number and inefficient or hard-to-optimize objective. To this end, we present a fast segmentation method for 3-D point clouds to reduce anchors, which can largely decrease the computing cost. Moreover, taking advantage of 3-D generalized Intersection of Union (GIoU) and L1 losses, we propose a fused loss to facilitate the optimization of 3-D object detection. A series of experiments show that the proposed method has alleviated the abovementioned issues effectively.

Aberration of Serum and Tissue N-Glycans in Mouse ß1,4-GalT1 Y286L Mutant Variants.

ß1,4-GalT1 is a type II membrane glycosyltransferase. It catalyzes the production of lactose in the lactating mammary gland and is supposedly also involved in the galactosylation of terminal GlcNAc of complex-type N-glycans. In-vitro studies of the bovine ß4Gal-T1 homolog showed that replacing a single residue of tyrosine with leucine at position 289 alters the donor substrate specificity from UDP-Gal to UDP-N-acetyl-galactosamine (UDP-GalNAc). The effect of this peculiar change in ß1,4GalT1 specificity was investigated in-vivo, by generating biallelic Tyr286Leu ß1,4GalT1 mice using CRISPR/Cas9 and crossbreeding. Mice bearing this mutation showed no appreciable defects when compared to wild-type mice, with the exception of biallelic female B4GALT1 mutant mice, which were unable to produce milk. The detailed comparison of wild-type and mutant mice derived from liver, kidney, spleen, and intestinal tissues showed only small differences in their N-glycan pattern. Comparable N-glycosylation was also observed in HEK 293 wild-type and knock-out B4GALT1 cells. Remarkably and in contrast to the other analyzed tissue samples, sialylation and galactosylation of serum N-glycans of biallelic Tyr286Leu GalT1 mice almost disappeared completely. These results suggest that ß1,4GalT1 plays a special role in the synthesis of serum N-glycans. The herein described Tyr286Leu ß1,4GalT1 mutant mouse model may, therefore, prove useful in the investigation of the mechanism which regulates tissue-dependent galactosylation.

Gluconeogenic growth of Vibrio cholerae is important for competing with host gut microbiota.

PURPOSE: The gastrointestinal tract is home to thousands of commensal bacterial species. Therefore, competition for nutrients is paramount for successful bacterial pathogen invasion of intestinal ecosystems. The human pathogen Vibrio cholerae, the causative agent of the severe diarrhoeal disease, cholera, is able to colonize the small intestine, which is protected by mucus. However, it is unclear which metabolic pathways or nutrients V. cholerae utilizes during intestinal colonization and growth. METHODOLOGY: In this study, we investigated the effect of various metabolic key genes, including those involved in the gluconeogenesis pathway, on V. cholerae physiology and in vivo colonization. RESULTS: We found that gluconeogenesis is important for infant mouse colonization. Growth assays showed that mutations in the key components of gluconeogenesis pathway, PpsA and PckA, lead to a growth defect in a minimal medium supplemented with mucin as a carbon source. Furthermore, the ppsA/pckA mutants colonized poorly in the adult mouse intestine, particularly when more gut commensal flora are present. CONCLUSION: Gluconeogenesis biosynthesis is important for the successful colonization of V. cholerae in a niche that is full of competing microbiota.

OxyR2 Modulates OxyR1 Activity and Vibrio cholerae Oxidative Stress Response.

Bacteria have developed capacities to deal with different stresses and adapt to different environmental niches. The human pathogen Vibrio cholerae, the causative agent of the severe diarrheal disease cholera, utilizes the transcriptional regulator OxyR to activate genes related to oxidative stress resistance, including peroxiredoxin PrxA, in response to hydrogen peroxide. In this study, we identified another OxyR homolog in V. cholerae, which we named OxyR2, and we renamed the previous OxyR OxyR1. We found that OxyR2 is required to activate its divergently transcribed gene ahpC, encoding an alkylhydroperoxide reductase, independently of H2O2 A conserved cysteine residue in OxyR2 is critical for this function. Mutation of either oxyR2 or ahpC rendered V. cholerae more resistant to H2O2 RNA sequencing analyses indicated that OxyR1-activated oxidative stress-resistant genes were highly expressed in oxyR2 mutants even in the absence of H2O2 Further genetic analyses suggest that OxyR2-activated AhpC modulates OxyR1 activity by maintaining low intracellular concentrations of H2O2 Furthermore, we showed that ΔoxyR2 and ΔahpC mutants were less fit when anaerobically grown bacteria were exposed to low levels of H2O2 or incubated in seawater. These results suggest that OxyR2 and AhpC play important roles in the V. cholerae oxidative stress response.