arXiv subjects

Hu Wang

Publications and source records attributed to Hu Wang.

At least 91 records · Page 5Linked to original sources

arXiv2020

Soft Expert Reward Learning for Vision-and-Language Navigation

Vision-and-Language Navigation (VLN) requires an agent to find a specified spot in an unseen environment by following natural language instructions. Dominant methods based on supervised learning clone expert's behaviours and thus perform better on seen environments, while showing restricted performance on unseen ones. Reinforcement Learning (RL) based models show better generalisation ability but have issues as well, requiring large amount of manual reward engineering is one of which. In this paper, we introduce a Soft Expert Reward Learning (SERL) model to overcome the reward engineering designing and generalisation problems of the VLN task. Our proposed method consists of two complementary components: Soft Expert Distillation (SED) module encourages agents to behave like an expert as much as possible, but in a soft fashion; Self Perceiving (SP) module targets at pushing the agent towards the final destination as fast as possible. Empirically, we evaluate our model on the VLN seen, unseen and test splits and the model outperforms the state-of-the-art methods on most of the evaluation metrics.

cs.CV↗

arXiv2020

Multi-label Thoracic Disease Image Classification with Cross-Attention Networks

Automated disease classification of radiology images has been emerging as a promising technique to support clinical diagnosis and treatment planning. Unlike generic image classification tasks, a real-world radiology image classification task is significantly more challenging as it is far more expensive to collect the training data where the labeled data is in nature multi-label; and more seriously samples from easy classes often dominate; training data is highly class-imbalanced problem exists in practice as well. To overcome these challenges, in this paper, we propose a novel scheme of Cross-Attention Networks (CAN) for automated thoracic disease classification from chest x-ray images, which can effectively excavate more meaningful representation from data to boost the performance through cross-attention by only image-level annotations. We also design a new loss function that beyond cross-entropy loss to help cross-attention process and is able to overcome the imbalance between classes and easy-dominated samples within each class. The proposed method achieves state-of-the-art results.

cs.CV↗

arXiv2020

Unsupervised Representation Learning by Predicting Random Distances

Deep neural networks have gained tremendous success in a broad range of machine learning tasks due to its remarkable capability to learn semantic-rich features from high-dimensional data. However, they often require large-scale labelled data to successfully learn such features, which significantly hinders their adaption into unsupervised learning tasks, such as anomaly detection and clustering, and limits their applications into critical domains where obtaining massive labelled data is prohibitively expensive. To enable unsupervised learning on those domains, in this work we propose to learn features without using any labelled data by training neural networks to predict data distances in a randomly projected space. Random mapping is a theoretically proven approach to obtain approximately preserved distances. To well predict these random distances, the representation learner is optimised to learn genuine class structures that are implicitly embedded in the randomly projected space. Empirical results on 19 real-world datasets show that our learned representations substantially outperform a few state-of-the-art competing methods in both anomaly detection and clustering tasks. Code is available at https://git.io/RDP

cs.CV↗

arXiv2019

A novel generative reverse net assisted evolution algorithm for expensive-computational optimizations

Simulation-based optimization is a useful method for practical design problems. However, it is difficult for complicated problems due to expensive-computational costs. A popular way to overcome this issue is to use a surrogate model to save the cost. Nevertheless, limited design parameters those are input to traditional surrogate models can difficultly represent the whole design problem, which might result in unexpected errors. In this study, physical cloud images from simulations are employed and attempted to construct the surrogate model. Simultaneously, based on the strong pattern recognition and generation abilities of deep learning models, a novel Generative Reverse Net assisted Evolution Algorithm (GRN-EA) is proposed for expensive-design problems. In this study, a numerical example of a Variable-Stiffness (VS) composite hole-plate is employed to obtain the optimal distribution of the curved fiber. Moreover, to evaluate the performance of GRN-EA in practical engineering problems, a more complex sheet-forming case is optimized. According to the two problems, some expensive-simulations can be well handled in this study.

eess.SP↗

arXiv2019

Topology optimization of heat sinks for instantaneous chip cooling using a transient pseudo-3D thermofluid model

With the increasing power density of electronics components, the heat dissipation capacity of heat sinks gradually becomes a bottleneck. Many structural optimization methods, including topology optimization, have been widely used for heat sinks. Due to its high design freedom, topology optimization is suggested for the design of heat sinks using a transient pseudo-3D thermofluid model to acquire better instantaneous thermal performance. The pseudo-3D model is designed to reduce the computational cost and maintain an acceptable accuracy. The model relies on an artificial heat convection coefficient to couple two layers and establish the approximate relationship with the corresponding 3D model. In the model, a constant pressure drop and heat generation rate are treated. The material distribution is optimized to reduce the average temperature of the base plate at the prescribed terminal time. Furthermore, to reduce the intermediate density regions during the density-based topology optimization procedure, a detailed analysis of interpolation functions is made and the penalty factors are chosen on this basis. Finally, considering the engineering application of the model, a practical model with more powerful cooling medium and higher inlet pressure is built. The optimized design shows a better instantaneous thermal performance and provides 66.7% of the pumping power reduction compared with reference design.

physics.flu-dyn↗

arXiv2019

Machine Learning based parameter tuning strategy for MMC based topology optimization

Moving Morphable Component (MMC) based topology optimization approach is an explicit algorithm since the boundary of the entity explicitly described by its functions. Compared with other pixel or node point-based algorithms, it is optimized through the parameter optimization of a Topological Description Function (TDF). However, the optimized results partly depend on the selection of related parameters of Method of Moving Asymptote (MMA), which is the optimizer of MMC based topology optimization. Practically, these parameters are tuned according to the experience and the feasible solution might not be easily obtained, even the solution might be infeasible due to improper parameter setting. In order to address these issues, a Machine Learning (ML) based parameter tuning strategy is proposed in this study. An Extra-Trees (ET) based image classifier is integrated to the optimization framework, and combined with Particle Swarm Optimization (PSO) algorithm to form a closed loop. It makes the optimization process be free from the manual parameter adjustment and the reasonable solution in the design domain is obtained. In this study, two classical cases are presented to demonstrate the efficiency of the proposed approach.

math.NA↗

arXiv2019

Variational Auto-Encoder Based Approximate Bayesian Computation Uncertian Inverse Method for Sheet Metal Forming Problem

In this study, an image-assisted Approximate Bayesian Computation (ABC) parameter inverse method is proposed to identify the design parameters. In the proposed method, the images are mapped to a low-dimensional latent space by Variational Auto-Encoder (VAE), and the information loss is minimized by network training. Therefore, an effective trade-off between information loss and computational cost can be achieved by using the latent variables of VAE as summary statistics of ABC, which overcomes the difficulty of selecting summary statistics in the ABC. Besides, for some practical engineering problems, processing the images as objective function can effective show the response result. Meanwhile, the relationship between design parameters and the latent variables is constructed by Least Squares Support Vector Regression (LSSVR) surrogate model. With the well-constructed LSSVR model, the simulation coefficient vectors under given parameters will be determined effectively. Then, the parameters to be identified are determined by comparing the simulated and observed coefficient vectors in ABC. Finally, a sheet forming problem is investgated by the suggested method. The material parameters of the blank and the process parameters of the forming process are identified. Results show that the method is feasibility and effective for the identification of sheet forming parameters.

eess.IV↗

arXiv2019

Image-based reconstruction for the impact problems by using DPNNs

With the improvement of the pattern recognition and feature extraction of Deep Neural Networks (DPNNs), image-based design and optimization have been widely used in multidisciplinary researches. Recently, a Reconstructive Neural Network (ReConNN) has been proposed to obtain an image-based model from an analysis-based model [1, 2], and a steady-state heat transfer of a heat sink has been successfully reconstructed. Commonly, this method is suitable to handle stable-state problems. However, it has difficulties handling nonlinear transient impact problems, due to the bottlenecks of the Deep Neural Network (DPNN). For example, nonlinear transient problems make it difficult for the Generative Adversarial Network (GAN) to generate various reasonable images. Therefore, in this study, an improved ReConNN method is proposed to address the mentioned weaknesses. Time-dependent ordered images can be generated. Furthermore, the improved method is successfully applied in impact simulation case and engineering experiment. Through the experiments, comparisons and analyses, the improved method is demonstrated to outperform the former one in terms of its accuracy, efficiency and costs.

cs.OH↗

arXiv2019

Can CNN Construct Highly Accurate Models Efficiently for High-Dimensional Problems in Complex Product Designs?

With the increase of the nonlinearity and dimension, it is difficult for the present popular metamodeling techniques to construct reliable metamodels. To address this problem, Convolutional Neural Network (CNN) is introduced to construct a highly accurate metamodel efficiently. Considering the inherent characteristics of the CNN, it is a potential modeling tool to handle highly nonlinear and dimensional problems (hundreds-dimensional problems) with the limited training samples. In order to evaluate the proposed CNN metamodel for hundreds-dimensional and strong nonlinear problems, CNN is compared with other metamodeling techniques. Furthermore, several high-dimensional analytical functions are also employed to test the CNN metamodel. Testing and comparisons confirm the efficiency and capability of the CNN metamodel for hundreds-dimensional and strong nonlinear problems. Moreover, the proposed CNN metamodel is also applied to IsoGeometric Analysis (IGA)-based optimization successfully.

cs.CV↗

arXiv2019

Image-Based Reconstruction for a 3D-PFHS Heat Transfer Problem by ReConNN

The heat transfer performance of Plate Fin Heat Sink (PFHS) has been investigated experimentally and extensively. Commonly, the objective function of the PFHS design is based on the responses of simulations. Compared with existing studies, the purpose of this study is to transfer from analysis-based model to image-based one for heat sink designs. Compared with the popular objective function based on maximum, mean, variance values etc., more information should be involved in image-based and thus a more objective model should be constructed. It means that the sequential optimization should be based on images instead of responses and more reasonable solutions should be obtained. Therefore, an image-based reconstruction model of a heat transfer process for a 3D-PFHS is established. Unlike image recognition, such procedure cannot be implemented by existing recognition algorithms (e.g. Convolutional Neural Network) directly. Therefore, a Reconstructive Neural Network (ReConNN), integrated supervised learning and unsupervised learning techniques, is suggested and improved to achieve higher accuracy. According to the experimental results, the heat transfer process can be observed more detailed and clearly, and the reconstructed results are meaningful for the further optimizations.

cs.CE↗

arXiv2018

Global sensitivity analysis based on DIRECT-KG-HDMR and thermal optimization of pin-fin heat sink for the platform inertial navigation system

In this study, in order to reduce the local high temperature of the platform in inertial navigation system (PINS), a pin-fin heat sink with staggered arrangement is designed. To reduce the dimension of the inputs and improve the efficiency of optimization, a feasible global sensitivity analysis (GSA) based on Kriging-High Dimensional Model Representation with DIviding RECTangles sampling strategy (DIRECT-KG-HDMR) is proposed. Compared with other GSA methods, the proposed method can indicate the effects of the structural and the material parameters on the maximum temperature at the bottom of the heat sink by using both sensitivity and coupling coefficients. From the results of GSA, it can be found that the structural parameters have greater effects on thermal performance than the material ones. Moreover, the coupling intensities between the structural and material parameters are weak. Therefore, the structural parameters are selected to optimize the thermal performance of the heat sink, and several popular optimization algorithms such as GA, DE, TLBO, PSO and EGO are used for the optimization. Moreover, steady thermal response of the PINS with the optimized heat sink is also studied, and its result shows that the maximum temperature of high temperature region of the platform is reduced by 1.09 degree Celsius compared with the PINS without the heat sink.

cs.CE↗

arXiv2018

IRA assisted MMC-based topology optimization method

An Iterative Reanalysis Approximation (IRA) is integrated with the Moving Morphable Components (MMCs) based topology optimization (IRA-MMC) in this study. Compared with other classical topology optimization methods, the Finite Element (FE) based solver is replaced with the suggested IRA method. In this way, the expensive computational cost can be significantly saved by several nested iterations. The optimization of linearly elastic planar structures is constructed by the MMC, the specifically geometric parameters of which are taken as design variables to acquire explicitly geometric boundary. In the suggested algorithm, a hybrid optimizer based on the Method of Moving Asymptotes (MMA) approach and the Globally Convergent version of the Method of Moving Asymptotes (GCMMA) is suggested to improve convergence ratio and avoid local optimum. The proposed approach is evaluated by some classical benchmark problems in topology optimization, where the results show significant time saving without compromising accuracy.

math.NA↗

arXiv2018

Closed loop image aided optimization for cold spray process based on molecular dynamics

This study proposed a closed loop image aided optimization (CLIAO) method to improve the quality of deposition during the cold spray process. Some recent research shows that the quality of deposition measured by flattening ratio of the bonded particle is associated with impact velocity, angle and particle size. Therefore, the original idea of CLIAO is to improve the quality of deposition by obtaining the maximum flattening ratio which is extracted from the molecular dynamics (MD) simulation snapshots directly. To complete this strategy, a Python script is suggested to generate the required snapshots from result files automatically and the image processing technique is used to evaluate the flattening ratio from the snapshots. Moreover, three optimization methods including surrogate optimization (Efficient Global Optimization) and heuristic algorithms (Particle Swarm Optimization, Different Evolution algorithm) are engaged. Then a back propagation neural network (BPNN) is used to accelerate the process of optimization, where the BPNN is used to build the meta-model instead of the forward calculation. The optimization result demonstrates that all the above methods can obtain the acceptable solution. The comparison between those methods is also given and the selection of them should be determined by the trade-off between efficiency and accuracy.

cs.CE↗

arXiv2018

Reconstruction of Simulation-Based Physical Field by Reconstruction Neural Network Method

A variety of modeling techniques have been developed in the past decade to reduce the computational expense and improve the accuracy of modeling. In this study, a new framework of modeling is suggested. Compared with other popular methods, a distinctive characteristic is "from image based model to analysis based model (e.g. stress, strain, and deformation)". In such a framework, a reconstruction neural network (ReConNN) model designed for simulation-based physical field's reconstruction is proposed. The ReConNN contains two submodels that are convolutional neural network (CNN) and generative adversarial net-work (GAN). The CNN is employed to construct the mapping between contour images of physical field and objective function. Subsequently, the GAN is utilized to generate more images which are similar to the existing contour images. Finally, Lagrange polynomial is applied to complete the reconstruction. However, the existing CNN models are commonly applied to the classification tasks, which seem to be difficult to handle with regression tasks of images. Meanwhile, the existing GAN architectures are insufficient to generate high-accuracy "pseudo contour images". Therefore, a ReConNN model based on a Convolution in Convolution (CIC) and a Convolutional AutoEncoder based on Wasserstein Generative Adversarial Network (WGAN-CAE) is suggested. To evaluate the performance of the proposed model representatively, a classical topology optimization procedure is considered. Then the ReConNN is utilized to the reconstruction of heat transfer process of a pin fin heat sink. It demonstrates that the proposed ReConNN model is proved to be a potential capability to reconstruct physical field for multidisciplinary, such as structural optimization.

cs.LG↗

arXiv2018

ReNN: Rule-embedded Neural Networks

The artificial neural network shows powerful ability of inference, but it is still criticized for lack of interpretability and prerequisite needs of big dataset. This paper proposes the Rule-embedded Neural Network (ReNN) to overcome the shortages. ReNN first makes local-based inferences to detect local patterns, and then uses rules based on domain knowledge about the local patterns to generate rule-modulated map. After that, ReNN makes global-based inferences that synthesizes the local patterns and the rule-modulated map. To solve the optimization problem caused by rules, we use a two-stage optimization strategy to train the ReNN model. By introducing rules into ReNN, we can strengthen traditional neural networks with long-term dependencies which are difficult to learn with limited empirical dataset, thus improving inference accuracy. The complexity of neural networks can be reduced since long-term dependencies are not modeled with neural connections, and thus the amount of data needed to optimize the neural networks can be reduced. Besides, inferences from ReNN can be analyzed with both local patterns and rules, and thus have better interpretability. In this paper, ReNN has been validated with a time-series detection problem.

cs.LG↗

arXiv2018

An Adaptive Population Size Differential Evolution with Novel Mutation Strategy for Constrained Optimization

Differential evolution (DE) has competitive performance on constrained optimization problems (COPs), which targets at searching for global optimal solution without violating the constraints. Generally, researchers pay more attention on avoiding violating the constraints than better objective function value. To achieve the aim of searching the feasible solutions accurately, an adaptive population size method and an adaptive mutation strategy are proposed in the paper. The adaptive population method is similar to a state switch which controls the exploring state and exploiting state according to the situation of feasible solution search. The novel mutation strategy is designed to enhance the effect of status switch based on adaptive population size, which is useful to reduce the constraint violations. Moreover, a mechanism based on multipopulation competition and a more precise method of constraint control are adopted in the proposed algorithm. The proposed differential evolution algorithm, APDE-NS, is evaluated on the benchmark problems from CEC2017 constrained real parameter optimization. The experimental results show the effectiveness of the proposed method is competitive compared to other state-of-the-art algorithms.

cs.NE↗

arXiv2018

A D-vine copula-based coupling uncertainty analysis for stiffness predication of variable-stiffness composite

This study suggests a coupling uncertainty analysis method to investigate the stiffness characteristics of variable stiffness (VS) composite. The D-vine copula function is used to address the coupling of random variables. To identify the copula relation between random variables, a novel one-step Bayesian copula model selection (OBCS) method is proposed to obtain a suitable copula function as well as the marginal CDF of random variables. The entire process is Monte Carlo simulation (MCS). However, due to the expensive computational cost of complete finite element analysis (FEA) in MCS, a fast solver, reanalysis method is introduced. To further improve the efficiency of entire procedure, a back propagation neural network (BPNN) model is also introduced based on the reanalysis method. Compared with the reanalysis method, BPNN shows a higher efficiency as well as sufficient accuracy. Finally, the fiber angle deviation of VS composite is investigated by the suggested strategy. Two numerical examples are presented to verify the feasibility of this method.

cs.CE↗

arXiv2017

A novel X-FEM based fast computational method for crack propagation

This study suggests a fast computational method for crack propagation, which is based on the extended finite element method (X-FEM). It is well known that the X-FEM might be the most popular numerical method for crack propagation. However, with the increase of complexity of the given problem, the size of FE model and the number of iterative steps are increased correspondingly. To improve the efficiency of X-FEM, an efficient computational method termed decomposed updating reanalysis (DUR) method is suggested. For most of X-FEM simulation procedures, the change of each iterative step is small and it will only lead a local change of stiffness matrix. Therefore, the DUR method is proposed to predict the modified response by only calculating the changed part of equilibrium equations. Compared with other fast computational methods, the distinctive characteristic of the proposed method is to update the modified stiffness matrix with a local updating strategy, which only the changed part of stiffness matrix needs to be updated. To verify the performance of the DUR method, several typical numerical examples have been analyzed and the results demonstrate that this method is a highly efficient method with high accuracy.

math.NA↗