ABSTRACT
We found that the area of black round or irregular-shaped spots on the tiger's nose increased with age, indicating a positive relationship between age and nose features. We used the deep learning model to train the facial and nose image features to identify the age of Amur tigers, using a combination of classification and prediction methods to achieve age determination with an accuracy of 87.81%.
ABSTRACT
Source-free domain adaptation (SFDA) shows the potential to improve the generalizability of deep learning-based face anti-spoofing (FAS) while preserving the privacy and security of sensitive human faces. However, existing SFDA methods are significantly degraded without accessing source data due to the inability to mitigate domain and identity bias in FAS. In this paper, we propose a novel Source-free Domain Adaptation framework for FAS (SDA-FAS) that systematically addresses the challenges of source model pre-training, source knowledge adaptation, and target data exploration under the source-free setting. Specifically, we develop a generalized method for source model pre-training that leverages a causality-inspired PatchMix data augmentation to diminish domain bias and designs the patch-wise contrastive loss to alleviate identity bias. For source knowledge adaptation, we propose a contrastive domain alignment module to align conditional distribution across domains with a theoretical equivalence to adaptation based on source data. Furthermore, target data exploration is achieved via self-supervised learning with patch shuffle augmentation to identify unseen attack types, which is ignored in existing SFDA methods. To our best knowledge, this paper provides the first full-stack privacy-preserving framework to address the generalization problem in FAS. Extensive experiments on nineteen cross-dataset scenarios show our framework considerably outperforms state-of-the-art methods.
ABSTRACT
3-D point clouds facilitate 3-D visual applications with detailed information of objects and scenes but bring about enormous challenges to design efficient compression technologies. The irregular signal statistics and high-order geometric structures of 3-D point clouds cannot be fully exploited by existing sparse representation and deep learning based point cloud attribute compression schemes and graph dictionary learning paradigms. In this paper, we propose a novel p-Laplacian embedding graph dictionary learning framework that jointly exploits the varying signal statistics and high-order geometric structures for 3-D point cloud attribute compression. The proposed framework formulates a nonconvex minimization constrained by p-Laplacian embedding regularization to learn a graph dictionary varying smoothly along the high-order geometric structures. An efficient alternating optimization paradigm is developed by harnessing ADMM to solve the nonconvex minimization. To our best knowledge, this paper proposes the first graph dictionary learning framework for point cloud compression. Furthermore, we devise an efficient layered compression scheme that integrates the proposed framework to exploit the correlations of 3-D point clouds in a structured fashion. Experimental results demonstrate that the proposed framework is superior to state-of-the-art transform-based methods in M-term approximation and point cloud attribute compression and outperforms recent MPEG G-PCC reference software.
ABSTRACT
By introducing randomness on the environments, domain randomization (DR) imposes diversity to the policy training of deep reinforcement learning, and thus improves its capability of generalization. The randomization of environments, however, introduces another source of variability for the estimate of policy gradients, in addition to the already high variance incurred by trajectory sampling. Therefore, with standard state-dependent baselines, the policy gradient methods may still suffer high variance, causing a low sample efficiency during the training of DR. In this paper, we theoretically derive a bias-free and state/environment-dependent optimal baseline for DR, and analytically show its ability to achieve further variance reduction over the standard constant and state-dependent baselines for DR. Based on our theory, we further propose a variance reduced domain randomization (VRDR) approach for policy gradient methods, to strike a tradeoff between the variance reduction and computational complexity for the practical implementation. By dividing the entire space of environments into some subspaces and then estimating the state/subspace-dependent baseline, VRDR enjoys a theoretical guarantee of variance reduction and faster convergence than the state-dependent baselines. Empirical evaluations on six robot control tasks with randomized dynamics demonstrate that VRDR not only accelerates the convergence of policy training, but can consistently achieve a better eventual policy with improved training stability.
ABSTRACT
Batch normalization (BN) is a fundamental unit in modern deep neural networks. However, BN and its variants focus on normalization statistics but neglect the recovery step that uses linear transformation to improve the capacity of fitting complex data distributions. In this paper, we demonstrate that the recovery step can be improved by aggregating the neighborhood of each neuron rather than just considering a single neuron. Specifically, we propose a simple yet effective method named batch normalization with enhanced linear transformation (BNET) to embed spatial contextual information and improve representation ability. BNET can be easily implemented using the depth-wise convolution and seamlessly transplanted into existing architectures with BN. To our best knowledge, BNET is the first attempt to enhance the recovery step for BN. Furthermore, BN is interpreted as a special case of BNET from both spatial and spectral views. Experimental results demonstrate that BNET achieves consistent performance gains based on various backbones in a wide range of visual tasks. Moreover, BNET can accelerate the convergence of network training and enhance spatial information by assigning important neurons with large weights accordingly.
ABSTRACT
It is promising to solve linear inverse problems by unfolding iterative algorithms (e.g., iterative shrinkage thresholding algorithm (ISTA)) as deep neural networks (DNNs) with learnable parameters. However, existing ISTA-based unfolded algorithms restrict the network architectures for iterative updates with the partial weight coupling structure to guarantee convergence. In this paper, we propose hybrid ISTA to unfold ISTA with both pre-computed and learned parameters by incorporating free-form DNNs (i.e., DNNs with arbitrary feasible and reasonable network architectures), while ensuring theoretical convergence. We first develop HCISTA to improve the efficiency and flexibility of classical ISTA (with pre-computed parameters) without compromising the convergence rate in theory. Furthermore, the DNN-based hybrid algorithm is generalized to popular variants of learned ISTA, dubbed HLISTA, to enable a free architecture of learned parameters with a guarantee of linear convergence. To our best knowledge, this paper is the first to provide a convergence-provable framework that enables free-form DNNs in ISTA-based unfolded algorithms. This framework is general to endow arbitrary DNNs for solving linear inverse problems with convergence guarantees. Extensive experiments demonstrate that hybrid ISTA can reduce the reconstruction error with an improved convergence rate in the tasks of sparse recovery and compressive sensing.
ABSTRACT
Endobronchial ultrasound (EBUS) elastography videos have shown great potential to supplement intrathoracic lymph node diagnosis. However, it is laborious and subjective for the specialists to select the representative frames from the tedious videos and make a diagnosis, and there lacks a framework for automatic representative frame selection and diagnosis. To this end, we propose a novel deep learning framework that achieves reliable diagnosis by explicitly selecting sparse representative frames and guaranteeing the invariance of diagnostic results to the permutations of video frames. Specifically, we develop a differentiable sparse graph attention mechanism that jointly considers frame-level features and the interactions across frames to select sparse representative frames and exclude disturbed frames. Furthermore, instead of adopting deep learning-based frame-level features, we introduce the normalized color histogram that considers the domain knowledge of EBUS elastography images and achieves superior performance. To our best knowledge, the proposed framework is the first to simultaneously achieve automatic representative frame selection and diagnosis with EBUS elastography videos. Experimental results demonstrate that it achieves an average accuracy of 81.29% and area under the receiver operating characteristic curve (AUC) of 0.8749 on the collected dataset of 727 EBUS elastography videos, which is comparable to the performance of the expert-based clinical methods based on manually-selected representative frames.
Subject(s)
Elasticity Imaging Techniques , Humans , Elasticity Imaging Techniques/methods , Thorax , Lymph Nodes/diagnostic imaging , Lymph Nodes/pathology , ROC Curve , Endosonography/methodsABSTRACT
Self-supervised learning based on instance discrimination has shown remarkable progress. In particular, contrastive learning, which regards each image as well as its augmentations as an individual class and tries to distinguish them from all other images, has been verified effective for representation learning. However, conventional contrastive learning does not model the relation between semantically similar samples explicitly. In this paper, we propose a general module that considers the semantic similarity among images. This is achieved by expanding the views generated by a single image to Cross-Samples and Multi-Levels, and modeling the invariance to semantically similar images in a hierarchical way. Specifically, the cross-samples are generated by a data mixing operation, which is constrained within samples that are semantically similar, while the multi-level samples are expanded at the intermediate layers of a network. In this way, the contrastive loss is extended to allow for multiple positives per anchor, and explicitly pulling semantically similar images together at different layers of the network. Our method, termed as CSML, has the ability to integrate multi-level representations across samples in a robust way. CSML is applicable to current contrastive based methods and consistently improves the performance. Notably, using MoCo v2 as an instantiation, CSML achieves 76.6% top-1 accuracy with linear evaluation using ResNet-50 as backbone, 66.7% and 75.1% top-1 accuracy with only 1% and 10% labels, respectively. All these numbers set the new state-of-the-art. The code is available at https://github.com/haohang96/CSML.
ABSTRACT
Industrial waste gas emissions from fossil fuel over-exploitation have aroused great attention in modern society. Recently, metal-organic frameworks (MOFs) have been developed in the capture and catalytic conversion of industrial exhaust gases such as SO2 , H2 S, NOx , CO2 , CO, etc. Based on these resourceful conversion applications, in this review, we summarize the crucial role of the surface, interface, and structure optimization of MOFs for performance enhancement. The main points include (1) adsorption enhancement of target molecules by surface functional modification, (2) promotion of catalytic reaction kinetics through enhanced coupling in interfaces, and (3) adaptive matching of guest molecules by structural and pore size modulation. We expect that this review will provide valuable references and illumination for the design and development of MOF and related materials with excellent exhaust gas treatment performance.
Subject(s)
Metal-Organic Frameworks , Industrial Waste , Adsorption , Catalysis , GasesABSTRACT
The Cox proportional hazards model is a popular semi-parametric model for survival analysis. In this paper, we aim at developing a federated algorithm for the Cox proportional hazards model over vertically partitioned data (i.e., data from the same patient are stored at different institutions). We propose a novel algorithm, namely VERTICOX, to obtain the global model parameters in a distributed fashion based on the Alternating Direction Method of Multipliers (ADMM) framework. The proposed model computes intermediary statistics and exchanges them to calculate the global model without collecting individual patient-level data. We demonstrate that our algorithm achieves equivalent accuracy for the estimation of model parameters and statistics to that of its centralized realization. The proposed algorithm converges linearly under the ADMM framework. Its computational complexity and communication costs are polynomially and linearly associated with the number of subjects, respectively. Experimental results show that VERTICOX can achieve accurate model parameter estimation to support federated survival analysis over vertically distributed data by saving bandwidth and avoiding exchange of information about individual patients. The source code for VERTICOX is available at: https://github.com/daiwenrui/VERTICOX.
ABSTRACT
Message passing has evolved as an effective tool for designing graph neural networks (GNNs). However, most existing methods for message passing simply sum or average all the neighboring features to update node representations. They are restricted by two problems: 1) lack of interpretability to identify node features significant to the prediction of GNNs and 2) feature overmixing that leads to the oversmoothing issue in capturing long-range dependencies and inability to handle graphs under heterophily or low homophily. In this article, we propose a node-level capsule graph neural network (NCGNN) to address these problems with an improved message passing scheme. Specifically, NCGNN represents nodes as groups of node-level capsules, in which each capsule extracts distinctive features of its corresponding node. For each node-level capsule, a novel dynamic routing procedure is developed to adaptively select appropriate capsules for aggregation from a subgraph identified by the designed graph filter. NCGNN aggregates only the advantageous capsules and restrains irrelevant messages to avoid overmixing features of interacting nodes. Therefore, it can relieve the oversmoothing issue and learn effective node representations over graphs with homophily or heterophily. Furthermore, our proposed message passing scheme is inherently interpretable and exempt from complex post hoc explanations, as the graph filter and the dynamic routing procedure identify a subset of node features that are most significant to the model prediction from the extracted subgraph. Extensive experiments on synthetic as well as real-world graphs demonstrate that NCGNN can well address the oversmoothing issue and produce better node representations for semisupervised node classification. It outperforms the state of the arts under both homophily and heterophily.
ABSTRACT
With the advent of data science, the analysis of network or graph data has become a very timely research problem. A variety of recent works have been proposed to generalize neural networks to graphs, either from a spectral graph theory or a spatial perspective. The majority of these works, however, focus on adapting the convolution operator to graph representation. At the same time, the pooling operator also plays an important role in distilling multiscale and hierarchical representations, but it has been mostly overlooked so far. In this article, we propose a parameter-free pooling operator, called iPool, that permits to retain the most informative features in arbitrary graphs. With the argument that informative nodes dominantly characterize graph signals, we propose a criterion to evaluate the amount of information of each node given its neighbors and theoretically demonstrate its relationship to neighborhood conditional entropy. This new criterion determines how nodes are selected and coarsened graphs are constructed in the pooling layer. The resulting hierarchical structure yields an effective isomorphism-invariant representation of networked data on arbitrary topologies. The proposed strategy achieves superior or competitive performance in graph classification on a collection of public graph benchmark data sets and superpixel-induced image graph data sets.
ABSTRACT
Model quantization is essential to deploy deep convolutional neural networks (DCNNs) on resource-constrained devices. In this article, we propose a general bitwidth assignment algorithm based on theoretical analysis for efficient layerwise weight and activation quantization of DCNNs. The proposed algorithm develops a prediction model to explicitly estimate the loss of classification accuracy led by weight quantization with a geometrical approach. Consequently, dynamic programming is adopted to achieve optimal bitwidth assignment on weights based on the estimated error. Furthermore, we optimize bitwidth assignment for activations by considering the signal-to-quantization-noise ratio (SQNR) between weight and activation quantization. The proposed algorithm is general to reveal the tradeoff between classification accuracy and model size for various network architectures. Extensive experiments demonstrate the efficacy of the proposed bitwidth assignment algorithm and the error rate prediction model. Furthermore, the proposed algorithm is shown to be well extended to object detection.
Subject(s)
Algorithms , Neural Networks, ComputerABSTRACT
BACKGROUND: Endoscopic ultrasound (EBUS) strain elastography can diagnose intrathoracic benign and malignant lymph nodes (LNs) by reflecting the relative stiffness of tissues. Due to strong subjectivity, it is difficult to give full play to the diagnostic efficiency of strain elastography. This study aims to use machine learning to automatically select high-quality and stable representative images from EBUS strain elastography videos. METHODS: LNs with qualified strain elastography videos from June 2019 to November 2019 were enrolled in the training and validation sets randomly at a quantity ratio of 3:1 to train an automatic image selection model using machine learning algorithm. The strain elastography videos in December 2019 were used as the test set, from which three representative images were selected for each LN by the model. Meanwhile, three experts and three trainees selected one representative image severally for each LN on the test set. Qualitative grading score and four quantitative methods were used to evaluate images above to assess the performance of the automatic image selection model. RESULTS: A total of 415 LNs were included in the training and validation sets and 91 LNs in the test set. Result of the qualitative grading score showed that there was no statistical difference between the three images selected by the machine learning model. Coefficient of variation (CV) values of the four quantitative methods in the machine learning group were all lower than the corresponding CV values in the expert and trainee groups, which demonstrated great stability of the machine learning model. Diagnostic performance analysis on the four quantitative methods showed that the diagnostic accuracies were range from 70.33% to 73.63% in the trainee group, 78.02% to 83.52% in the machine learning group, and 80.22% to 82.42% in the expert group. Moreover, there were no statistical differences in corresponding mean values of the four quantitative methods between the machine learning and expert groups (p >0.05). CONCLUSION: The automatic image selection model established in this study can help select stable and high-quality representative images from EBUS strain elastography videos, which has great potential in the diagnosis of intrathoracic LNs.
ABSTRACT
HIGHLIGHTS: Hard-carbon anode dominated with ultra-micropores (< 0.5 nm) was synthesized for sodium-ion batteries via a molten diffusion-carbonization method. The ultra-micropores dominated carbon anode displays an enhanced capacity, which originates from the extra sodium-ion storage sites of the designed ultra-micropores. The thick electrode (~ 19 mg cm-2) with a high areal capacity of 6.14 mAh cm-2 displays an ultrahigh cycling stability and an outstanding low-temperature performance. Pore structure of hard carbon has a fundamental influence on the electrochemical properties in sodium-ion batteries (SIBs). Ultra-micropores (< 0.5 nm) of hard carbon can function as ionic sieves to reduce the diffusion of slovated Na+ but allow the entrance of naked Na+ into the pores, which can reduce the interficial contact between the electrolyte and the inner pores without sacrificing the fast diffusion kinetics. Herein, a molten diffusion-carbonization method is proposed to transform the micropores (> 1 nm) inside carbon into ultra-micropores (< 0.5 nm). Consequently, the designed carbon anode displays an enhanced capacity of 346 mAh g-1 at 30 mA g-1 with a high ICE value of ~ 80.6% and most of the capacity (~ 90%) is below 1 V. Moreover, the high-loading electrode (~ 19 mg cm-2) exhibits a good temperature endurance with a high areal capacity of 6.14 mAh cm-2 at 25 °C and 5.32 mAh cm-2 at - 20 °C. Based on the in situ X-ray diffraction and ex situ solid-state nuclear magnetic resonance results, the designed ultra-micropores provide the extra Na+ storage sites, which mainly contributes to the enhanced capacity. This proposed strategy shows a good potential for the development of high-performance SIBs.
ABSTRACT
Three-dimensional (3D) TiO2 architectures have attracted significant attention recently as they can improve the electrochemical stability and realize the full potential of TiO2-based anodes in lithium ion batteries. Here, flower-like rutile TiO2 spheres with radially assembled nanorods (c-channels) were fabricated via a simple hydrothermal method. The 3D radial architecture affords massive active sites to fortify the lithium storage. Moreover, the presence of c-channels facilitates electrolyte infiltration and offers facile pathways for efficient Li+ transport. As a result, this flower-like rutile TiO2 anode gives significantly enhanced specific capacities (615 mA h g-1 at 1 C and 386 mA h g-1 at 2 C after 400 cycles) and a superior long-term cyclability (up to 10 000 cycles with a specific capacity of 67 mA h g-1 at 100 C). Kinetic analysis reveals that the enhanced diffusion-controlled and surface capacitive storage leads to the excellent electrochemical behavior. This work not only exhibits the enormous advantages of 3D architectures with c-channels, but also provides access to structural design and crystal phase selection for TiO2-based anode materials.
ABSTRACT
Lithium-oxygen batteries with ultrahigh energy densities have drawn considerable attention as next-generation energy storage devices. However, their practical applications are challenged by sluggish reaction kinetics aimed at the formation/decomposition of discharge products on battery cathodes. Developing effective catalysts and understanding the fundamental catalytic mechanism are vital to improve the electrochemical performance of lithium-oxygen batteries. Here, uniformly dispersed ruthenium nanoparticles anchored on nitrogen-doped reduced graphene oxide are prepared by using an in situ pyrolysis procedure as a bifunctional catalyst for lithium-oxygen batteries. The abundance of ruthenium active sites and strong ruthenium-support interaction enable a feasible discharge product formation/decomposition route by modulating the surface adsorption of lithium superoxide intermediates and the nucleation and growth of lithium peroxide species. Benefiting from these merits, the electrode provides a drastically increased discharge capacity (17,074 mA h g-1), a decreased charge overpotential (0.51 V), and a long-term cyclability (100 cycles at 100 mA g-1). Our observations reveal the significance of the dispersion and coordination of metal catalysts, shedding light on the rational design of efficient catalysts for future lithium-oxygen batteries.
ABSTRACT
Photocatalytic hydrogen peroxide (H2O2) generation represents a promising approach for artificial photosynthesis. However, the sluggish half-reaction of water oxidation significantly limits the efficiency of H2O2 generation. Here, a benzylamine oxidation with more favorable thermodynamics is employed as the half-reaction to couple with H2O2 generation in water by using defective zirconium trisulfide (ZrS3) nanobelts as a photocatalyst. The ZrS3 nanobelts with disulfide (S22-) and sulfide anion (S2-) vacancies exhibit an excellent photocatalytic performance for H2O2 generation and simultaneous oxidation of benzylamine to benzonitrile with a high selectivity of >99%. More importantly, the S22- and S2- vacancies can be separately introduced into ZrS3 nanobelts in a controlled manner. The S22- vacancies are further revealed to facilitate the separation of photogenerated charge carriers. The S2- vacancies can significantly improve the electron conduction, hole extraction, and kinetics of benzylamine oxidation. As a result, the use of defective ZrS3 nanobelts yields a high production rate of 78.1 ± 1.5 and 32.0 ± 1.2 µmol h-1 for H2O2 and benzonitrile, respectively, under a simulated sunlight irradiation.
ABSTRACT
Metallic nanostructures are commonly densely packed into a few packing variants with slightly different atomic packing factors. The structural aspects and physicochemical properties related with the vacancies in such nanostructures are rarely explored because of lack of an effective way to control the introduction of vacancy sites. Highly voided metallic nanostructures with ordered vacancies are however energetically high lying and very difficult to synthesize. Here, we report a chemical method for synthesis of hierarchical Rh nanostructures (Rh NSs) composed of ultrathin nanosheets, composed of hexagonal close-packed structure embedded with nanodomains that adopt a vacated Barlow packing with ordered vacancies. The obtained Rh NSs exhibit remarkably enhanced electrocatalytic activity and stability toward the hydrogen evolution reaction (HER) in alkaline media. Theoretical calculations reveal that the exceptional electrocatalytic performance of Rh NSs originates from their unique vacancy structures, which facilitate the adsorption and dissociation of H2O in the HER.
ABSTRACT
Differentiable architecture search (DARTS) enables effective neural architecture search (NAS) using gradient descent, but suffers from high memory and computational costs. In this paper, we propose a novel approach, namely Partially-Connected DARTS (PC-DARTS), to achieve efficient and stable neural architecture search by reducing the channel and spatial redundancies of the super-network. In the channel level, partial channel connection is presented to randomly sample a small subset of channels for operation selection to accelerate the search process and suppress the over-fitting of the super-network. Side operation is introduced for bypassing (non-sampled) channels to guarantee the performance of searched architectures under extremely low sampling rates. In the spatial level, input features are down-sampled to eliminate spatial redundancy and enhance the efficiency of the mixed computation for operation selection. Furthermore, edge normalization is developed to maintain the consistency of edge selection based on channel sampling with the architectural parameters for edges. Theoretical analysis shows that partial channel connection and parameterized side operation are equivalent to regularizing the super-network on the weights and architectural parameters during bilevel optimization. Experimental results demonstrate that the proposed approach achieves higher search speed and training stability than DARTS. PC-DARTS obtains a top-1 error rate of 2.55 percent on CIFAR-10 with 0.07 GPU-days for architecture search, and a state-of-the-art top-1 error rate of 24.1 percent on ImageNet (under the mobile setting) within 2.8 GPU-days.