Convolutional Long Short-term Memory Network Research Articles

Portfolio management based on a reinforcement learning framework

AbstractPortfolio management is crucial for investors. We propose a dynamic portfolio management framework based on reinforcement learning using the proximal policy optimization algorithm. The two‐part framework includes a feature extraction network and a full connected network. First, the majority of the previous research on portfolio management based on reinforcement learning has been dedicated to discrete action spaces. We propose a potential solution to the problem of a continuous action space with a constraint (i.e., the sum of the portfolio weights is equal to 1). Second, we explore different feature extraction networks (i.e., convolutional neural network [CNN], long short‐term memory [LSTM] network, and convolutional LSTM network) combined with our system, and we conduct extensive experiments on the six kinds of assets, including 16 features. The empirical results show that the CNN performs best in the test set. Last, we discuss the effect of the trading frequency on our trading system and find that the monthly trading frequency has a higher Sharpe ratio in the test set than other trading frequencies.

Short-term multi-step-ahead sector-based traffic flow prediction based on the attention-enhanced graph convolutional LSTM network (AGC-LSTM)

AbstractAccurate sector-based air traffic flow predictions are essential for ensuring the safety and efficiency of the air traffic management (ATM) system. However, due to the inherent spatial and temporal dependencies of air traffic flow, it is still a challenging problem. To solve this problem, some methods are proposed considering the relationship between sectors, while the complicated spatiotemporal dynamics and interdependencies between traffic flow of route segments related to the sector are not taken into account. To address this challenge, the attention-enhanced graph convolutional long short-term memory network (AGC-LSTM) model is applied to improve the short-term sector-based traffic flow prediction, in which spatial structures of route segments related to the sector are considered for the first time. Specifically, the graph convolutional networks (GCN)-LSTM network model was employed to capture spatiotemporal dependencies of the flight data, and the attention mechanism is designed to concentrate on the informative features from key nodes at each layer of the AGC-LSTM model. The proposed model is evaluated through a case study of the typical enroute sector in the central–southern region of China. The prediction results show that MAE reduces by 14.4% compared to the best performing GCN-LSTM model among the other five models. Furthermore, the study involves comparative analyses to assess the influence of route segment range, input and output sequence lengths, and time granularities on prediction performance. This study helps air traffic managers predict flight situations more accurately and avoid implementing overly conservative or excessively aggressive flow management measures for the sectors.

A Prolonged Artificial Nighttime-light Dataset of China (1984-2020)

Nighttime light remote sensing has been an increasingly important proxy for human activities. Despite an urgent need for long-term products and pilot explorations in synthesizing them, the publicly available long-term products are limited. A Night-Time Light convolutional LSTM network is proposed and applied the network to produce a 1-km annual Prolonged Artificial Nighttime-light DAtaset of China (PANDA-China) from 1984 to 2020. Assessments between modeled and original images show that on average the RMSE reaches 0.73, the coefficient of determination (R2) reaches 0.95, and the linear slope is 0.99 at the pixel level, indicating a high confidence in the quality of generated data products. Quantitative and visual comparisons witness PANDA-China’s superiority against other NTL datasets in its significantly longer NTL dynamics, higher temporal consistency, and better correlations with socioeconomics (built-up areas, gross domestic product, population) characterizing the most relevant indicator in different development phases. The PANDA-China product provides an unprecedented opportunity to trace nighttime light dynamics in the past four decades.

The Prediction of the Remaining Useful Life of Rotating Machinery Based on an Adaptive Maximum Second-Order Cyclostationarity Blind Deconvolution and a Convolutional LSTM Autoencoder.

The prediction of the remaining useful life (RUL) is important for the conditions of rotating machinery to maintain reliability and decrease losses. This study proposes an efficient approach based on an adaptive maximum second-order cyclostationarity blind deconvolution (ACYCBD) and a convolutional LSTM autoencoder to achieve the feature extraction, health index analysis, and RUL prediction for rotating machinery. First, the ACYCBD is used to filter noise from the vibration signals. Second, based on the peak value properties, a novel health index (HI) is designed to analyze the health conditions for the denoising signal, showing a high sensitivity for the degradation of bearings. Finally, for better prognostics and health management of the rotating machinery, based on convolutional layers and LSTM, an autoencoder can achieve a transform convolutional LSTM network to develop a convolutional LSTM autoencoder (ALSTM) model that can be applied to forecast the health trend for rotating machinery. Compared with the SVM, CNN, LSTM, GRU, and DTGRU methods, our experiments demonstrate that the proposed approach has the greatest performance for the prediction of the remaining useful life of rotating machinery.

Quantum Convolutional Long Short-Term Memory Based on Variational Quantum Algorithms in the Era of NISQ

In the era of noisy intermediate-scale quantum (NISQ) computing, the synergistic collaboration between quantum and classical computing models has emerged as a promising solution for tackling complex computational challenges. Long short-term memory (LSTM), as a popular network for modeling sequential data, has been widely acknowledged for its effectiveness. However, with the increasing demand for data and spatial feature extraction, the training cost of LSTM exhibits exponential growth. In this study, we propose the quantum convolutional long short-term memory (QConvLSTM) model. By ingeniously integrating classical convolutional LSTM (ConvLSTM) networks and quantum variational algorithms, we leverage the variational quantum properties and the accelerating characteristics of quantum states to optimize the model training process. Experimental validation demonstrates that, compared to various LSTM variants, our proposed QConvLSTM model outperforms in terms of performance. Additionally, we adopt a hierarchical tree-like circuit design philosophy to enhance the model’s parallel computing capabilities while reducing dependence on quantum bit counts and circuit depth. Moreover, the inherent noise resilience in variational quantum algorithms makes this model more suitable for spatiotemporal sequence modeling tasks on NISQ devices.

A novel adaptive deep transfer learning method towards thermal error modeling of electric spindles under variable conditions

Thermal error modeling (TEM) plays a vital role in maintaining the machining accuracy of electric spindles. Recently, deep learning (DL) techniques have obtained promising achievements in this area. However, DL techniques have certain limitations. The data acquired from variable working conditions present large distribution discrepancies, the DL-based model established on one working condition fails to obtain satisfactory prediction accuracy in another working condition. Moreover, existing studies focus solely on using temperature features to build prediction models, neglecting the full use of multi-sensory information. To address these issues, this paper proposes a novel adaptive deep transfer learning method towards TEM of electric spindles, which takes full advantage of the temperature, current, and power sensory information. Firstly, finite element simulation is employed to analyze the thermal characteristics of the electric spindle and determine the locations for temperature measurement points. Then, the convolutional long short-term memory network (C-LSTMN) is constructed, where the spatial features from multi-sensory information between inputs and prediction patterns are captured by convolutional layers, these features are further processed by long short-term memory network (LSTMN) to extract temporal features. Subsequently, the multi-kernel joint maximum mean discrepancy (MK-JMMD) measure is developed to minimize the distribution discrepancies between the source and target domains, thus the prediction model initially established on the source domain can be adaptive to effectively predict on the target domain. Finally, with unavailable thermal error data in the target domain, the proposed method is validated through 12 transfer tasks using datasets from four working conditions with other comparison methods. The results demonstrate that the proposed method overcomes the challenges of unavailable labeled thermal error samples and outperforms advanced methods.

Deep Convolutional LSTM for improved flash flood prediction

Flooding remains one of the most devastating and costly natural disasters. As flooding events grow in frequency and intensity, it has become increasingly important to improve flood monitoring, prediction, and early warning systems. Recent efforts to improve flash flood forecasts using deep learning have shown promise, yet commonly-used techniques such as long short term memory (LSTM) models are unable to extract potentially significant spatial relationships among input datasets. Here we propose a hybrid approach using a Convolutional LSTM (ConvLSTM) network to predict stream stage heights using multi-modal hydrometeorological remote sensing and in-situ inputs. Results suggest the hybrid network can more effectively capture the specific spatiotemporal landscape dynamics of a flash flood-prone catchment relative to the current state-of-the-art, leading to a roughly 26% improvement in model error when predicting elevated stream conditions. Furthermore, the methodology shows promise for improving prediction accuracy and warning times for supporting local decision making.

A novel model for fall detection and action recognition combined lightweight 3D-CNN and convolutional LSTM networks

A novel model for fall detection and action recognition combined lightweight 3D-CNN and convolutional LSTM networks

Forecasting solar irradiation using convolutional long short-term memory and feature selection of data from neighboring locations

The capacity of renewable energy systems, such as photovoltaic farms, has been steadily increasing owing to the continuous depletion of fossil-based energy sources. Information that can be used to optimize the use of renewable energy is required. This work proposes a novel method that uses data from neighboring locations to perform 24 h-ahead solar GHI forecasting. The approach is helpful in supporting the operation of photovoltaic farms, minimizing the need to collect new data on the target location. This work proposes the Convolutional Long Short-term Memory (ConvLSTM) model to maximize the performance using a feature-rich spatiotemporal dataset. The ConvLSTM model uses Hadamard convolutions in its input-to-state transition and state-to-state transition, allowing it to retain better information. The feature selection methods are used to ensure that only the critical subsets of the dataset are used to develop the model. The subsets identified by the feature selection methods are used on Vector Autoregressive, Support Vector Regression, Multilayer Perceptron, Convolutional Neural Network (CNN), CNN-LSTM and ConvLSTM networks. The winning method is compared to related studies using Stacked-LSTM with transfer-learning, attention-based LSTM and Krill-Herd Algorithm – Support vector Regression (KHA-SVR). The resulting coefficients of determination and root-mean-squared errors were statistically evaluated and compared. Ultimately, the proposed ConvLSTM model performed consistently better across all the seasons and feature subsets.

Differential Diagnosis of Benign and Malignant Pulmonary Nodules in CT Images Based on Multitask Learning.

Artificial intelligence-based aided diagnostic systems for pulmonary nodules can be divided into subtasks such as nodule detection, segmentation, and benign and malignant differentiation. Most current studies are limited to single-target tasks. However, aided diagnosis aims to distinguish benign from malignant pulmonary nodules, which requires the fusion of multiple-scale features and comprehensive discrimination based on the results of multiple learning tasks. This study focuses on the aspects of model design, network structure, and constraints and proposes a novel model that integrates the learning tasks of pulmonary nodule detection, segmentation, and classification under weakly supervised conditions. The main innovations include the following three aspects: (1) a two-dimensional sequence detection model based on a ConvLSTM (Convolutional Long Short-Term Memory) network and U-shaped structure network is proposed to obtain the context space features of image slices fully; (2) a differential diagnosis of benign and malignant pulmonary nodules based on multitask learning is proposed, which uses the annotated data of different types of tasks to mine the potential common features among tasks; and (3) an optimization strategy incorporating prior knowledge of computed tomography images and dynamic weight adjustment of multiple tasks is proposed to ensure that each task can efficiently complete training and learning. Experiments on the LIDC-IDRI and LUNA16 datasets showed that our proposed method achieved a final competition performance metric score of 87.80% for nodule detection and a Dice similarity coefficient score of 83.95% for pulmonary nodule segmentation. The cross-validation results of the LIDC-IDRI and LUNA16 datasets show that our model achieved 87.80% of the final competition performance metric score for nodule detection and 83.95% of the DSC score for pulmonary nodule segmentation, representing the optimal result for that dataset.

An Unsupervised Convolutional LSTM Network (C-LSTMNet) for Lung 4D-CT Registration

An Unsupervised Convolutional LSTM Network (C-LSTMNet) for Lung 4D-CT Registration

Comparison of model selection and data bias on the prediction performance of purpleback flying squid (Sthenoteuthis oualaniensis) fishing ground in the Northwest Indian Ocean

Purpleback flying squid (Sthenoteuthis oualaniensis, PFS) is one of the critical economically cephalopod species in the northwest Indian Ocean, and the accurate model and dataset selection are critical parameters for predicting and managing the PFS fishing grounds. In this study, the PFS fishery data for the years 2016–2021 was analyzed using the gravity center of fishing grounds method, generalized additive model (GAM), gradient boosted trees (GBT), 3D Convolutional Neural Network (3DCNN), and 3D Convolutional Neural Networks-Convolutional LSTM Network (3DCNN-ConvLSTM) to explore the differences in annual catches, the annual gravity center of fishing ground, model performance, and importance of environmental variables in the case of datasets A (no moonlight days) and B (no moonlight days + bright moonlight days). The results are as follows: 1) Datasets A and B exhibit similar patterns of variation, with annual catches rising and then declining and the annual gravity center of fishing ground moving northeastward overall; 2) The GAM and GBT models had better model performance on dataset B (GAM (average F1-score ± standard deviation): 0.678658 ± 0.014684; GBT: 0.737422 ± 0.011748)than on dataset A (GAM: 0.676802 ± 0.013403; GBT: 0.736547 ± 0.013323), but almost negligible, and the standard deviation of the GAM model on dataset B becomes larger. 3DCNN and 3DCNN-ConvLSTM models perform in contrast, with significantly better F1-score and standard deviations on dataset A (3DCNN: 0.75048 ± 0.019763; 3DCNN-ConvLSTM: 0.740041 ± 0.023927) than on dataset B (3DCNN: 0.746378 ± 0.020337; 3DCNN-ConvLSTM: 0.736927 ± 0.04498); 3) The 3DCNN (optimal prediction of performance) or GBT (Optimal model stability) model is optimal for predicting the PFS fishing grounds; 4) GBT, 3DCNN, and 3DCNN-ConvLSTM model results all showed that the differences in the importance of environmental variables obtained from datasets A and B were significant; 5) Unlike the GBT model, the 3DCNN and 3DCNN-ConvLSTM models were more susceptible to dataset influences, with significant dependence on environmental variables that have large positive and negative sample differences. This study provides rich suggestions for constructing a predictive model for the PFS fishing grounds in the context of climate change. It also provides a new perspective on cleaning up biased data for light fisheries.

Short-term prediction of tropical cyclone track and intensity via four mainstream deep learning techniques

Tropical cyclone (TC) is a highly destructive natural disaster, whose impact is closely correlated to its track and intensity. Thus, it is crucial to obtain the information on TC track and intensity both timely and accurately for early warning and mitigation of related disasters. While there are numerous prediction methods, deep learning (DL) technology has demonstrated exceptional potential in recent years. However, studies on TC prediction via DL techniques are still insufficient, and in particular there is a lack of efforts toward the dependence of prediction results upon some key parameters involved in the prediction strategy. To this end, this paper present a comprehensive study on TC prediction via four mainstream DL techniques, i.e., Long Short-Term Memory (LSTM) network, Convolutional LSTM (CovnLSTM) network, Temporal Convolutional Network (TCN), and the recently developed Transformer model. The performance of different DL techniques is examined, with a highlight on the effects of three setting parameters involved in the prediction process on the model performance, i.e., length of input records Nin, interval between input records ΔT, and lead time of the prediction Nout. Results reveal that TCN performs best from the perspectives of prediction accuracy, running efficiency and stability, followed by ConvLSTM; whilst LSTM and Transformer fail to provide competitive predictions, although Transformer showcases notable training efficiency. In reference to the prediction strategy, increasing ΔT and Nout can consistently enlarge prediction errors across all models, while the model performance can be optimized by setting the value of Nin in a range of 3–6. It is interesting to find that the overall performance of all models can be improved noticeably by using records with a finer ΔT in the scenarios with the same lead time but with varied Nout. Meanwhile, TCN can be further improved by extending Nout from Nout = m to Nout = n (n > m) and using the m-th step prediction in the scenario with Nout = n as the output in the scenario with Nout = m. The presented results provide useful insights for reasonably utilizing DL models to deal with prediction issues, which are applicable not only for TCs but also for other objectives that are widely concerned in wind engineering, e.g., wind speed/pressure and structural responses.

Retracted: Classification of Alzheimer's Disease Using Gaussian-Based Bayesian Parameter Optimization for Deep Convolutional LSTM Network.

[This retracts the article DOI: 10.1155/2021/4186666.].

Daily-Scale Prediction of Arctic Sea Ice Concentration Based on Recurrent Neural Network Models

Arctic sea ice prediction is of great practical significance in facilitating Arctic route planning, optimizing fisheries management, and advancing the field of sea ice dynamics research. While various deep learning models have been developed for sea ice prediction, they predominantly operate at the seasonal or sub-seasonal scale, often focusing on localized areas, and few cater to full-region daily-scale prediction. This study introduces the use of spatiotemporal sequence data prediction models, namely, the convolutional LSTM (ConvLSTM) and predictive recurrent neural network (PredRNN), for the prediction of sea ice concentration (SIC). Our analysis reveals that, when solely utilizing SIC historical data as the input, the ConvLSTM model outperforms the PredRNN model in SIC prediction. To enhance the models’ capacity to capture spatiotemporal relationships between multiple variables, we expanded the range of input data types to form the ConvLSTM-multi and PredRNN-multi models. Experimental findings demonstrate that the prediction accuracy of the four models significantly surpasses the CMIP6 model in three prospective climate scenarios (SSP126, SSP245, and SSP585). Of the four models, the ConvLSTM-multi model excels in assimilating the influence of reanalysis data on sea ice within the sea ice edge region, thus exhibiting superior performance than the PredRNN-multi model in predicting daily Arctic SIC over the subsequent 10 days. Furthermore, sensitivity tests on various model parameters highlight the substantial impact of sea surface temperature and prediction date on the accuracy of daily sea ice prediction, and meteorological and oceanographic parameters primarily affect the prediction accuracy of the thin-ice region at the edge of the sea ice.

Machinery Prognostics and High-Dimensional Data Feature Extraction Based on a Transformer Self-Attention Transfer Network.

Machinery degradation assessment can offer meaningful prognosis and health management information. Although numerous machine prediction models based on artificial intelligence have emerged in recent years, they still face a series of challenges: (1) Many models continue to rely on manual feature extraction. (2) Deep learning models still struggle with long sequence prediction tasks. (3) Health indicators are inefficient for remaining useful life (RUL) prediction with cross-operational environments when dealing with high-dimensional datasets as inputs. This research proposes a health indicator construction methodology based on a transformer self-attention transfer network (TSTN). This methodology can directly deal with the high-dimensional raw dataset and keep all the information without missing when the signals are taken as the input of the diagnosis and prognosis model. First, we design an encoder with a long-term and short-term self-attention mechanism to capture crucial time-varying information from a high-dimensional dataset. Second, we propose an estimator that can map the embedding from the encoder output to the estimated degradation trends. Then, we present a domain discriminator to extract invariant features from different machine operating conditions. Case studies were carried out using the FEMTO-ST bearing dataset, and the Monte Carlo method was employed for RUL prediction during the degradation process. When compared to other established techniques such as the RNN-based RUL prediction method, convolutional LSTM network, Bi-directional LSTM network with attention mechanism, and the traditional RUL prediction method based on vibration frequency anomaly detection and survival time ratio, our proposed TSTN method demonstrates superior RUL prediction accuracy with a notable SCORE of 0.4017. These results underscore the significant advantages and potential of the TSTN approach over other state-of-the-art techniques.

ECG Signal Classification with Convolutional Neural Networks

Cardiovascular waves in electrocardiograms (ECG) give significant data on heart conditions and the impacts of heart drugs. Division and analysis of ECG and its constituent cardiovascular waves are of high importance in cardiology determination and drug studies. Customarily, thoroughly prepared clinicians and cardiologists have performed ECG investigations. In any case, using clinicians in enormous scope ECG screening, for example, drug test stages or populace-based screening programs is not monetarily achievable. Consequently, an automated ECG division approach that can fragment heart waves accurately is of high significance. This thesis concentrates on Deep Learning (DL) based automated ECG division and outline techniques. Because of different shapes and irregularities in the ECG signal, the traditional straightforward component channels neglect to remove the assortment of cardiovascular wave arrangements. Convolutional Neural Networks (CNN) apply multi-facet including channels on the input to remove complex features from the sign. In this way, CNN can be used to remove various levelled highlights from ECG signals. Two Convnet structures are read up and utilized for the restriction of cardiovascular waves. Their exhibitions are contrasted with one another and different investigations in the writing too. The outcome shows that CNN is capable of extracting heart wave spatial elements with the great execution. Moreover, various long haul notwithstanding transient worldly patterns exists in ECG signals because of arrhythmia and other heart conditions. Generally speaking, momentary memory will be unable to catch the transient highlights in ECG signals, thus a Long Short-Term Memory (LSTM) network is planned and utilized to catch long haul and momentary information dependencies in ECG sequences. This technique has worked on distinguishing proof of worldly highlights, especially in P-wave. To address the crude ECG signal division issue, a hybrid DL compressing of convolutional auto encoders and LSTM networks is proposed and contemplated. Results show that it performs commonly well for crude ECG signal division.

Improving Cancer Classification Using Deep Reinforcement Learning with Convolutional LSTM Networks

Improving Cancer Classification Using Deep Reinforcement Learning with Convolutional LSTM Networks

ConvST-LSTM-Net: convolutional spatiotemporal LSTM networks for skeleton-based human action recognition

ConvST-LSTM-Net: convolutional spatiotemporal LSTM networks for skeleton-based human action recognition

An active learning convolutional neural network for predicting river flow in a human impacted system

The South Platte river system contains a mixture of natural streams, reservoirs, and pipeline projects that redirect water to front range communities in Colorado. At many timepoints, a simple persistence model is the best predictor for flow from pipelines and reservoirs but at other times, flows change based on snowmelt and inputs such as reservoir fill rates, local weather, and anticipated demand. Here we find that a convolutional Long Short-Term Memory (LSTM) network is well suited to modeling flow in parts of this basin that are strongly impacted by water projects as well as ones that are relatively free from direct human modifications. Furthermore, it is found that including an active learning component in which separate Convolutional Neural Networks (CNNs) are used to classify and then select the data that is then used for training a convolutional LSTM network is advantageous. Models specific for each gauge are created by transfer of parameter from a base model and these gauge-specific models are then fine-tuned based a curated subset of training data. The result is accurate predictions for both natural flow and human influenced flow using only past river flow, reservoir capacity, and historical temperature data. In 14 of the 16 gauges modeled, the error in the prediction is reduced when using the combination of on-the-fly classification by CNN followed by analysis by either a persistence or convolutional LSTM model. The methods designed here could be applied broadly to other basins and to other situations where multiple models are needed to fit data at different times and locations.