Multi-information-based convolutional neural network with attention mechanism for pedestrian trajectory prediction

Non-probability sampling network based on anomaly pedestrian trajectory discrimination for pedestrian trajectory prediction

paper, we focus on exploring the relationship between weave patterns and their mechanical properties in woven fiber composites through Machine Learning. Specifically, we explore the interactions between woven architectures and in-plane stiffness properties through Deep Convolutional Neural Network (DCNN) and Generative Adversarial Network (GAN). Our research is important for exploring how woven composite’s pattern is related to its mechanical properties and accelerating woven composite design as well as optimization. We focus on two tasks: (1) Stiffness prediction: Predicting in-plane stiffness properties for given weave patterns. Our DCNN extracts high-level features through several convolutional and fully connected layers to determine the final predictions. (2) Weave pattern prediction: Predicting weave patterns for target stiffness properties, which can be treated as the reverse task of the first one. Due to many-to-one mapping between weave patterns and the composite properties, we utilize a Decoder Neural Network as our baseline model and compare its performance with GAN and Genetic Algorithm. We represent the weave patterns as 2D checkerboard models and use finite element analysis (FEA) to determine in-plane stiffness properties, which serve as input data for our ML framework. We show that: (1) for stiffness prediction, DCNN can predict stiffness values for a given weave pattern with relatively high accuracy (above 93%); (2) for weave pattern prediction, the GAN model gives the best prediction accuracy (above 92%) while Decoder Neural Network has the best time efficiency. HAOTIAN FENG

Application of machine learning algorithms in the domain of financial engineering

Future pedestrian trajectory prediction offers great prospects for many practical applications such as unmanned vehicles, building evacuation design and robotic path planning. Most existing methods focus on social interaction among pedestrians but ignore the fact that heterogeneous traffic objects (cars, dogs, bicycles, motorcycles, etc.) have significant influence on the future trajectory of a subject pedestrian. Also, the walking direction intention of a pedestrian may be referred by his/her facial keypoints. Considering this, this work proposes to predict a pedestrian's future trajectory by jointly using neighboring heterogeneous traffic information and his/her facial keypoints. To fulfill this, an end-to-end facial keypoints-based convolutional encoder-decoder network (FK-CEN) is designed, in which the heterogeneous traffic and facial keypoints are input. After training, FK-CEN is evaluated on 5 crowded video sequences collected from the public datasets MOT-16 and MOT-17. Experimental results demonstrate that it outperforms state-of-the-art approaches, in terms of prediction errors.

Prediction of PM2.5 concentration in urban agglomeration of China by hybrid network model

In order to accurately predict the lane‐changing trajectory of the vehicle and improve the driving safety of the vehicle, a lane‐changing trajectory prediction model based on the combination of convolutional neural network (CNN) and bidirectional long short‐term memory (BiLSTM) neural network is proposed by comprehensively considering the historical driving behavior, the spatial characteristics of surrounding vehicles and the bidirectional time sequence information of the vehicle trajectory. Firstly, the vehicle trajectory data are filtered and smoothed, and it is divided into three categories: left lane change, right lane change, and straight driving, and a lane change trajectory sample set is constructed. Secondly, CNN‐BiLSTM model is constructed to identify the sample set of lane‐changing trajectory. Considering the interaction between vehicles in the driving process, the information of predicted vehicle, and surrounding vehicles is taken as the input of the model. The extracted feature vector is input to the BiLSTM layer for prediction after the CNN layer feature extraction, and the horizontal and vertical coordinates of the target vehicle at the next time are output. Thirdly, the trajectory data of the US‐101 dataset in NGSIM is selected to verify the performance of the CNN‐BiLSTM model, and at the same time, it is compared with models such as CNN‐LSTM, long short‐term memory (LSTM), BiLSTM, and CNN‐GRU‐Att. Finally, the verification result shows that the overall fitting degree of the vehicle lane change trajectory prediction of the proposed model reaches 99.50%, and the mean square error and mean absolute error are 0.0003076 and 0.01417, which are improved compared with other models. In the meanwhile, the research on multistep trajectory prediction in different prediction time domains is carried out. It was found that the longer the prediction time domain is, the lower the prediction performance of the model decreases, but the prediction accuracy still reached more than 96%, and it was able to accurately predict the lane change trajectory.

Reservoir characterization through seismic data analysis is essential for exploration and production in the petroleum industry. However, seismic-to-well tie discrepancies, limited availability of high-quality well data, and resolution constraints pose a reliability challenge. While previous studies offer valuable insights, they still struggle to achieve high-resolution predictions in a complex geologically environment given high reliance on well data. This study integrates synthetic data-driven techniques with real data, including convolutional neural networks (CNN) and transfer learning, to improve seismic reservoir characterization. We utilize nearby well statistics and a rock physics model (RPM) to simulate pseudo wells representing various geological scenarios. Synthetic seismic gathers are generated from these pseudo wells, which are based on RPM and local well control, to train the CNN. Transfer learning is then applied to adapt the CNN to better distinguish between real and synthetic data, enhancing reservoir predictions. A comparative analysis of P-impedance predictions from three methodologies: theory-driven Pre-Stack-Seismic-Inversion (TDSI), Deep-Neural-Network (DNN), and our CNN approach, showed that CNN achieved nearly 97% prediction accuracy with low error rates, compared to relatively lower prediction accuracy rates of DNN (86.2%) and TDSI (81.5%) with high error rates, according to robust metrics including R-square, RMSE, MSE, and MAE. These results indicate that CNN not only enhanced resolution but also closely aligned with well data and superior lateral continuity, even in blind well scenarios. This study effectively integrates synthetic data-driven techniques with CNNs and transfer learning to advance seismic reservoir property prediction, offering a robust solution to overcome limitations in traditional and DNN-based approaches.

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 203962, “Upscaling of Realistic Discrete Fracture Simulations Using Machine Learning,” by Nikolai Andrianov, SPE, Geological Survey of Denmark and Greenland, prepared for the 2021 SPE Reservoir Simulation Conference, Galveston, Texas, 4–6 October. The paper has not been peer reviewed. Upscaling of discrete fracture networks to continuum models such as the dual-porosity/dual-permeability (DP/DP) model is an industry-standard approach in modeling fractured reservoirs. In the complete paper, the author parametrizes the fine-scale fracture geometries and assesses the accuracy of several convolutional neural networks (CNNs) to learn the mapping between this parametrization and DP/DP model closures. The accuracy of the DP/DP results with the predicted model closures was assessed by a comparison with the corresponding fine-scale discrete fracture matrix (DFM) simulation of a two-phase flow in a realistic fracture geometry. The DP/DP results matched the DFM reference solution well. The DP/DP model also was significantly faster than DFM simulation. Introduction The goal of this study was to evaluate the effect of different CNN architectures on prediction accuracy for the DP/DP model closures and on the accuracy of DP/DP simulations in comparison with fine-scale DFM simulations. As a starting point, two CNN configurations were considered that have achieved breakthrough performance in image-classification tasks. The author adopted these architectures to the problem of learning the mapping between pixelated fracture geometries and the DP/DP model closures and indicated several key features in the CNN structure that are crucial for achieving high prediction accuracy. Mapping of fracture geometries requires significant effort, which limits the possibilities for creating large training data sets with realistic fracture geometries. The author, therefore, used the synthetic random linear fractures’ data set to train the CNNs and the fracture geometry from the Lägerdorf outcrop for testing purposes. It was demonstrated that an optimal CNN configuration yielded the DP/DP model closures such that the corresponding DP/DP results matched well the two-phase DFM simulations on a subset of the Lägerdorf data. The run times for the DP/DP model were a fraction of the time needed to accomplish the DFM simulations. Problem formulation is presented in a series of equations in the complete paper.

Traffic flow forecasting is an important problem for the successful deployment of intelligent transportation systems, which has been studied for more than two decades. In recent years, deep learning methods are emerging to serve as the benchmark tool for traffic flow forecasting due to its superior prediction performance. However, most studies are based on simple deep learning methods that can not capture inter- and intra-day traffic patterns as well as the correlation between contextual factors like the weather and the traffic flow. In this paper, we propose a novel deep-learning-based method for daily traffic flow forecasting where incorporating contextual factors and traffic flow patterns can be critical. First, a particular convolutional neural network (CNN) is deployed to extract inter- and intra- day traffic flow patterns. Then extracted features are fed into long short-term memory (LSTM) units to learn the intra-day temporal evolution of traffic flow. Finally, contextual information of historical days is integrated to enhance the prediction performance. Through a real-data case study, we show that the proposed approach achieves over 90% prediction accuracy which greatly outperforms existing benchmark methods and its forecasting performance is robust under various scenarios.

Although RGBD cameras can provide depth information in real scenes, the captured depth map is often of low resolution and insufficient quality compared to the color image. Typically, most of the existing methods work by assuming that the edges in depth map and its corresponding color image are more likely to occur simultaneously. However, when the color image is rich in detail, the high-frequency information which is non-existent in the depth map will be introduced into the depth map. In this paper, we propose a CNN-based method to detect the co-occurrent structural edge for color-guided depth map super-resolution. Firstly, we design an edge detection convolutional neural network (CNN) to obtain the co-occurrent structural edge in depth map and its corresponding color image. Then we pack the obtained co-occurrent structural edges and the interpolated low-resolution depth maps into another customized CNN for depth map super-resolution. The presented scheme can effectively interpret and exploit the structural correlation between the depth map and the color image. Additionally, recursive learning is adopted to reduce the parameters of the customized CNN for depth map super-resolution and avoid overfitting. Experimental results demonstrate the effectiveness and reliability of our proposed approach by comparing with the state-of-the-art methods.

Tsunamis are among the most catastrophic natural disasters, often triggered by seismic events such as earthquakes. Accurately predicting tsunami occurrences based on seismic parameters is critical for mitigating their devastating impacts. This study investigates the application of three advanced deep learning architectures such as Convolutional Neural Networks (CNNs), Long Short-Term Memory Networks (LSTMs), and Gated Recurrent Units (GRUs) for binary classification of tsunami events using seismic data. The dataset comprises earthquake records from 1995 to 2023, including features such as magnitude, depth, latitude, longitude, Modified Mercalli Intensity (MMI), and Community Internet Intensity (CDI). The models were evaluated using stratified 10-fold cross-validation and assessed across precision, recall, F1-score, accuracy, and ROC-AUC metrics. Results indicate that CNN outperformed the other architectures, achieving the highest accuracy (72.5%), precision (0.5987), and ROC-AUC (0.7838). GRU demonstrated moderate performance, balancing computational efficiency and predictive accuracy with an accuracy of 71.7% and ROC-AUC of 0.7709. LSTM, while theoretically adept at modeling temporal dependencies, showed the lowest performance due to challenges in capturing the dataset’s characteristics. The findings emphasize the importance of selecting architecture suited to the dataset’s features and task requirements. CNN’s superior performance highlights its effectiveness in spatial pattern extraction, while GRU offers a computationally efficient alternative. Future work will explore hybrid models and the integration of additional features to enhance prediction robustness. This study contributes to advancing tsunami prediction methodologies, supporting early warning systems for disaster preparedness.

This paper presents a novel framework for accurate pedestrian trajectory prediction in intersection corners or near crosswalks. Given prior knowledge of curbside geometry (i.e. angle made by intersecting curbs at the corner point of interest and the coordinates of the corner itself), the presented framework can accurately predict pedestrian trajectories even in new, unseen intersections. This is achieved by learning motion primitives in a common frame, called the curbside coordinate frame. A key insight in developing this common frame is to ensure that trajectories from intersections with different geometries, representing the same behavior, are spatially similar in the common frame. Motion primitives learned in such a common frame, can then be easily generalized to predict in new intersections, with different geometries than the ones trained on. We test our algorithm on real pedestrian trajectory datasets collected at two intersections, with distinctly different curbside and crosswalk geometries. A comparison of our algorithm with [1] demonstrates improved prediction accuracies of pedestrian trajectory prediction in the case of same training and test intersections, and the improvement of accuracy in the most different training and test intersections scenarios. The result also shows additional context, such as information about pedestrian traffic lights, if available, can be easily incorporated in our prediction model for further improvement in prediction accuracy.

Stock price prediction plays a crucial role in financial decision-making, especially in volatile market conditions. Traditional forecasting methods, including statistical models and deep learning approaches, often struggle to adapt to rapid market fluctuations and unexpected events such as the COVID-19 pandemic. Moreover, many existing models require large datasets and fail to generalize well across different financial conditions. Addressing these challenges, generative models have gained attention for their ability to generate synthetic financial data and improve prediction accuracy. Here, we propose a Wasserstein Generative Adversarial Network (WGAN) with Gated Recurrent Units (GRU) as the generator and Convolutional Neural Networks (CNN) as the discriminator to enhance stock price forecasting robustness. Our model is trained using historical stock prices, trading volume, and technical indicators from four major U.S. companies—Apple, Amazon, Google, and Intel—over a period from 2014 to 2023. The dataset is split into pre-and post-COVID-19 periods to assess the model's adaptability to market anomalies. Experimental results indicate that the proposed WGAN-based model achieves an average Root Mean Square Error (RMSE) of 2.31, outperforming conventional methods such as Linear Regression and Long Short-Term Memory (LSTM) networks. Moreover, the model demonstrates minimal performance degradation when exposed to volatile periods, proving its robustness in handling market uncertainties. These findings suggest that WGAN with GRU and CNN is an effective approach to mitigating the impact of market volatility on stock price prediction. Future research will explore further optimizations, alternative generative architectures, and applications to different financial markets to refine prediction accuracy and generalizability.

Analyzing the microscopic dynamics of pushing behavior within crowds can offer valuable insights into crowd patterns and interactions. By identifying instances of pushing in crowd videos, a deeper understanding of when, where, and why such behavior occurs can be achieved. This knowledge is crucial to creating more effective crowd management strategies, optimizing crowd flow, and enhancing overall crowd experiences. However, manually identifying pushing behavior at the microscopic level is challenging, and the existing automatic approaches cannot detect such microscopic behavior. Thus, this article introduces a novel automatic framework for identifying pushing in videos of crowds on a microscopic level. The framework comprises two main components: (i) feature extraction and (ii) video detection. In the feature extraction component, a new Voronoi-based method is developed for determining the local regions associated with each person in the input video. Subsequently, these regions are fed into EfficientNetV1B0 Convolutional Neural Network to extract the deep features of each person over time. In the second component, a combination of a fully connected layer with a Sigmoid activation function is employed to analyze these deep features and annotate the individuals involved in pushing within the video. The framework is trained and evaluated on a new dataset created using six real-world experiments, including their corresponding ground truths. The experimental findings demonstrate that the proposed framework outperforms state-of-the-art approaches, as well as seven baseline methods used for comparative analysis.

The prediction of stock price movement direction is significant in financial circles and academic. Stock price contains complex, incomplete, and fuzzy information which makes it an extremely difficult task to predict its development trend. Predicting and analysing financial data is a nonlinear, time-dependent problem. With rapid development in machine learning and deep learning, this task can be performed more effectively by a purposely designed network. This paper aims to improve prediction accuracy and minimizing forecasting error loss through deep learning architecture by using Generative Adversarial Networks. It was proposed a generic model consisting of Phase-space Reconstruction (PSR) method for reconstructing price series and Generative Adversarial Network (GAN) which is a combination of two neural networks which are Long Short-Term Memory (LSTM) as Generative model and Convolutional Neural Network (CNN) as Discriminative model for adversarial training to forecast the stock market. LSTM will generate new instances based on historical basic indicators information and then CNN will estimate whether the data is predicted by LSTM or is real. It was found that the Generative Adversarial Network (GAN) has performed well on the enhanced root mean square error to LSTM, as it was 4.35% more accurate in predicting the direction and reduced processing time and RMSE by 78 secs and 0.029, respectively. This study provides a better result in the accuracy of the stock index. It seems that the proposed system concentrates on minimizing the root mean square error and processing time and improving the direction prediction accuracy, and provides a better result in the accuracy of the stock index.