Generative Adversarial Networks:Introduction and Outlook

Recently, generative adversarial networks(GANs)have become a research focus of artificial intelligence. Inspired by two-player zero-sum game, GANs comprise a generator and a discriminator, both trained under the adversarial learning idea.The goal of GANs is to estimate the potential distribution of real data samples and generate new samples from that distribution.Since their initiation, GANs have been widely studied due to their enormous prospect for applications, including image and vision computing, speech and language processing, etc. In this review paper, we summarize the state of the art of GANs and look into the future. Firstly, we survey GANs' proposal background,theoretic and implementation models, and application fields.Then, we discuss GANs' advantages and disadvantages, and their development trends. In particular, we investigate the relation between GANs and parallel intelligence,with the conclusion that GANs have a great potential in parallel systems research in terms of virtual-real interaction and integration. Clearly, GANs can provide substantial algorithmic support for parallel intelligence.

Training and Testing Object Detectors With Virtual Images

In the area of computer vision, deep learning has produced a variety of state-of-the-art models that rely on massive labeled data. However, collecting and annotating images from the real world is too demanding in terms of labor and money investments, and is usually inflexible to build datasets with specific characteristics, such as small area of objects and high occlusion level. Under the framework of Parallel Vision, this paper presents a purposeful way to design artificial scenes and automatically generate virtual images with precise annotations.A virtual dataset named Parallel Eye is built, which can be used for several computer vision tasks. Then, by training the DPM(Deformable parts model) and Faster R-CNN detectors, we prove that the performance of models can be significantly improved by combining Parallel Eye with publicly available real-world datasets during the training phase. In addition, we investigate the potential of testing the trained models from a specific aspect using intentionally designed virtual datasets, in order to discover the flaws of trained models. From the experimental results, we conclude that our virtual dataset is viable to train and test the object detectors.

A Recurrent Attention and Interaction Model for Pedestrian Trajectory Prediction

The movement of pedestrians involves temporal continuity,spatial interactivity,and random diversity.As a result,pedestrian trajectory prediction is rather challenging.Most existing trajectory prediction methods tend to focus on just one aspect of these challenges,ignoring the temporal information of the trajectory and making too many assumptions.In this paper,we propose a recurrent attention and interaction(RAI)model to predict pedestrian trajectories.The RAI model consists of a temporal attention module,spatial pooling module,and randomness modeling module.The temporal attention module is proposed to assign different weights to the input sequence of a target,and reduce the speed deviation of different pedestrians.The spatial pooling module is proposed to model not only the social information of neighbors in historical frames,but also the intention of neighbors in the current time.The randomness modeling module is proposed to model the uncertainty and diversity of trajectories by introducing random noise.We conduct extensive experiments on several public datasets.The results demonstrate that our method outperforms many that are state-ofthe-art.

Prediction of methane storage in covalent organic frameworks using big-data-mining approach

A combination of computational materials screening and machine learning(ML)technique is being adopted as a popular approach to study various materials toward application of interest.In this work,we began with high-throughput molecular simulations to calculate the methane storage(6.5 MPa)and deliverable(6.5-0.58 MPa)capacities of 404,460 covalent organic frameworks(COFs)at 298 K.Then,the full data sets with 23 features were randomly split into training and test sets in a ratio of 20:80,which were applied to evaluate the prediction abilities of several ML algorithms,including gradient boosting decision tree(GBDT),neural network(NN),support vector machine(SVM),random forest(RF)and decision tree(DT).The results indicate that the RF model has the highest prediction accuracy,which was further employed to reduce the dimension of features space and quantitatively analyze the relative importance of each feature value.The binary classification predictors built using the features with the highest influence weight can give a successful identification of top-performing candidates from the test set containing 323,168 COFs with an accuracy exceeding 96%.The deliverable capacities of the identified COFs were found to outperform those reported so far for various adsorbents.The findings may provide a useful guidance for the design and synthesis of new high-performance materials for methane storage application.

A decouple-decomposition noise analysis model for closed-loop mode-localized tilt sensors

The development of mode-localized sensors based on amplitude output metrics has attracted increasing attention in recent years due to the potential of such sensors for high sensitivity and resolution.Mode-localization phenomena leverage the interaction between multiple coupled resonant modes to achieve enhanced performance,providing a promising solution to overcome the limitations of traditional sensing technologies.Amplitude noise plays a key role in determining the resolution of mode-localized sensors,as the output metric is derived from the measured AR(amplitude ratio)within the weakly coupled resonator system.However,the amplitude noise originating from the weakly coupled resonator's closed-loop circuit has not yet been fully investigated.This paper presents a decouple-decomposition(DD)noise analysis model,which is applied to achieve high resolution in a mode-localized tilt sensor based on a weakly coupled resonator closed-loop circuit.The DD noise model separates the weakly coupled resonators using the decoupling method considering the nonlinearity of the resonators.By integrating the decoupled weakly coupled resonators,the model decomposes the weakly coupled resonator's closed-loop circuit into distinct paths for amplitude and phase noise analyses.The DD noise model reveals noise effects at various circuit nodes and models the system noise in the closed-loop circuit of the weakly coupled resonators.MATLAB/Simulink simulations verify the model's accuracy when compared to theoretical analysis.At the optimal operating point,the mode-localized tilt sensor achieves an input-referred instability of 3.91×10^(-4°)and an input-referred AR of PSD of 2.01×10^(-4°)⁄√Hz using the closed-loop noise model.This model is also applicable to other varieties of mode-localized sensors.