Enhanced Outbound Volume Prediction for Refined Oil Depots via an Innovative Multitask Learning Framework

A unified modeling framework for lane change intention recognition and vehicle status prediction

A data-driven hybrid framework for voltage transformer ratio error prediction: Addressing challenges in complex power systems

ADMET property prediction via multi-task graph learning under adaptive auxiliary task selection

Trajectory regression, which aims to predict the travel time of arbitrary trajectories on road networks, attracts significant attention in various applications of traffic systems these years. In this paper, we tackle this problem with a multitask learning (MTL) framework. To take the temporal nature of the problem into consideration, we divide the regression problem into a set of sub-tasks of distinct time periods, then the problem can be treated in a multi-task learning framework. Further, we propose a novel regularization term in which we exploit the block sparse structure to augment the robustness of the model. In addition, we incorporate the spatial smoothness over road links and thus achieve a spatial-temporal framework. An accelerated proximal algorithm is adopted to solve the convex but non-smooth problem, which will converge to the global optimum. Experiments on both synthetic and real data sets demonstrate the effectiveness of the proposed method.

In this work we utilize implied volatility predictions from options on individual securities to predict the average volatility of an ensemble of securities. We cast this as a problem of online learning with expert advice, which has been studied extensively in the Machine Learning community. In this setting we treat each security as an expert and adaptively maintain a set of weights to combine the implied volatility predictions of all securities in our ensemble. We also consider options of different term lengths, treating each term length as a separate learning task in a Multi-task Learning (MTL) framework. In this MTL framework information is shared between the learning tasks to improve the overall learning process. We conduct experiments using historical option pricing data and corresponding historical volatility records. Using the straight mean of the implied volatility predictions as a benchmark strategy, we demonstrate improved combined predictions with a basic online learning algorithm and further improvement with our MTL framework.

In this paper, based on the widely available GPS trajectory big data that records the driving behaviours implicitly, we propose a multi-task learning (MTL) framework called \textsf{semi-Traj2Graph} to recognize the fine-grained driving styles in the temporal dimension accurately. The MTL framework can incorporate the learning capability of graph representation in extracting high-level and interpretable features regarding complex driving behaviours and semi-supervised in exploiting unlabelled data and reducing labelling effort. More specifically, in the graph representation learning, a multi-view graph is first built to capture a more complete view of driving behaviours from the raw GPS trajectory data, then graph convolutional neural networks (Graph-CNNs) are applied. In the semi-supervised learning, a pseudo-label labelling is adopted to make use of the unlabelled data. We evaluate the proposed framework extensively based on two taxi trajectory datasets collected from the city of Beijing and Chongqing, China, respectively. Experimental results show that \textsf{semi-Traj2Graph} outperforms compared to other baselines, achieving an overall accuracy of around 90\%. We also implement the framework on users' smartphones via the collaborative cloud-edge computation manner to demonstrate the system usability in real case

Objective. Deep learning (DL)-based automated contouring and treatment planning has been proven to improve the efficiency and accuracy of radiotherapy. However, conventional radiotherapy treatment planning process has the automated contouring and treatment planning as separate tasks. Moreover in DL, the contouring and dose prediction tasks for automated treatment planning are done independently.Approach. In this study, we applied the multi-task learning (MTL) approach in order to seamlessly integrate automated contouring and voxel-based dose prediction tasks, as MTL can leverage common information between the two tasks and be able to increase the efficiency of the automated tasks. We developed our MTL framework using the two datasets: in-house prostate cancer dataset and the publicly available head and neck cancer dataset, OpenKBP.Main results. Compared to the sequential DL contouring and treatment planning tasks, our proposed method using MTL improved the mean absolute difference of dose volume histogram metrics of prostate and head and neck sites by 19.82% and 16.33%, respectively. Our MTL model for automated contouring and dose prediction tasks demonstrated enhanced dose prediction performance while maintaining or sometimes even improving the contouring accuracy. Compared to the baseline automated contouring model with the Dice score coefficients of 0.818 for prostate and 0.674 for head and neck datasets, our MTL approach achieved average scores of 0.824 and 0.716 for these datasets, respectively.Significance. Our study highlights the potential of the proposed automated contouring and planning using MTL to support the development of efficient and accurate automated treatment planning for radiotherapy.

Healthcare professionals and individual users use wearable devices equipped with various sensors for healthcare management. Recently, the joint usage of artificial intelligence and these wearable sensors has played an essential role in healthcare management by providing a wide range of applications such as fitness tracking, gym activity monitoring, patient rehabilitation monitoring, and disease detection. These tasks eventually aim to enhance personal well-being and better manage the user’s physical health by monitoring different activity types and body weight changes. Here, we present an efficient multi-task learning framework based on commercial smart insoles that can solve three tasks related to physical health management: activity classification, speed estimation, and body weight estimation. Our multi-task framework converts the sensor data from the smart insole to a recurrence plot, which shows significant performance improvement compared to processing the raw time series data. In addition, we utilized a modified MobileNetV2 as our backbone network, which has a total parameter of less than 100K and a computational budget of 0.34G of multiply-accumulate operations. Furthermore, we collected a vast dataset from 72 users carrying out 16 experiments, which contains the largest number of people for multi-task learning purposes using smart insoles. Extensive experiments show that the proposed multi-task learning framework is extremely efficient while outperforming or leading to comparable performance against single-task models.

A self-adaptive loss physics-informed neural network (SAL-PINN) method is proposed to address the challenges of limited generalization capability exhibited by purely data-driven neural networks under sparse data conditions and the insufficient accuracy of standard physics-informed neural networks in capturing complex flow details. This method enables high-fidelity reconstruction of the plane turbine cascade flow field with limited measurement points. By incorporating a dynamic weight balancing mechanism, SAL-PINN adaptively adjusts the loss function weights for regions characterized by different numerical scales within the flow field, thereby facilitating targeted reinforcement learning for critical regions. This approach yields excellent performance in predicting both global flow structures and local flow details. The results demonstrate that under sparse data conditions, SAL-PINN achieves a further reduction in prediction errors for key physical quantities across the entire field by approximately 1.5%–2.0% compared to the standard physics-informed neural networks. In boundary regions with complex flows, the error reduction is even more pronounced, reaching 12.0%–25.4%. This study also systematically evaluates the models' sensitivity to data volume, revealing that the incorporation of physical information significantly reduces their dependence on data, while the adaptive weighting mechanism further diminishes this sensitivity. This work provides an effective solution for achieving high-accuracy flow field reconstruction with minimal data, offering significant potential for engineering applications, particularly in scenarios where experimental measurements are high-cost.

Due to the recent vast availability of transportation traffic data, major research efforts have been devoted to traffic prediction, which is useful in many applications such as urban planning, traffic management and navigations systems. Current prediction methods that independently train a model per traffic sensor cannot accurately predict traffic in every situation (e.g., rush hours, constructions and accidents) because there may not exist sufficient training samples per sensor for all situations. To address this shortcoming, our core idea is to explore the commonalities of prediction tasks across multiple sensors who behave similarly in a specific traffic situation. Instead of building a model independently per sensor, we propose a Multi-Task Learning (MTL) framework that aims to first automatically identify the traffic situations and then simultaneously build one forecasting model for similar-behaving sensors per traffic situation. The key innovation here is that instead of the straightforward application of MTL where each "task" corresponds to a sensor, we relate each MTL's "task" to a traffic situation. Specifically, we first identify these traffic situations by running clustering algorithms on all sensors' data. Subsequently, to enforce the commonalities under each identified situation, we use the group Lasso regularization in MTL to select a common set of features for the prediction tasks, and we adapt efficient FISTA algorithm with guaranteed convergence rate. We evaluated our methods with a large volume of real-world traffic sensor data; our results show that by incorporating traffic situations, our proposed MTL framework performs consistently better than naively applying MTL per sensor. Moreover, our holistic approach, under different traffic situations, outperforms all the best traffic prediction approaches for a given situation by up to 18% and 30% in short and long term predictions, respectively.

High-resolution LAI data can provide more detailed information for remote sensing applications. However, for deep learning-based LAI retrieval methods, a training dataset composed of sensor data with a moderate or low spatial resolution cannot be directly applied to derive LAI from high-spatial-resolution sensor data due to distribution discrepancies between different sensor data. Following in this insight, we proposed a multi-task adversarial transfer learning framework, named DANN_MT. This framework leverages an adaptive adversarial training strategy to extract domain-invariant features between MODIS and Landsat surface reflectance data, enabling the retrieval of high-resolution LAI. To address the incomplete consistency between tasks of generating corresponding moderate-resolution LAI and high-resolution LAI, the DANN_MT model is designed with multi-task learning. Comparison results suggest that the retrieved high-resolution LAI exhibit reasonable spatial distribution and seasonality. Direct validation demonstrates that the retrieved high-resolution LAI (RMSE for the GBOV sites: 0.86, RMSE for the IMAGINES sites: 0.82) have a higher accuracy than the SNAP LAI (RMSE for the GBOV sites: 0.93, RMSE for the IMAGINES sites: 0.95). These conclusions reveal that the knowledge learned from the moderate-spatial-resolution dataset is successfully transferred to produce the high-spatial-resolution LAI from Landsat surface reflectance using the proposed DANN_MT model.

Assessing the performance of physics-informed neural networks for tumor growth prediction under noisy and sparse data conditions.

Effectively predicting interactive behaviors of traffic participants in the urban road is the key to successful decision-making and motion planning of intelligent vehicles. In this article, based on the data collected from vehicle on-board sensors, a graph-neural-network-based multitask learning framework (GNN-MTLF) is proposed to accurately predict trajectories of traffic participants with interactive behaviors. The interactive behavior considered in this research includes interactive events and trajectories that are modeled as spatial-temporal graphs using the GNN. Under the GNN-MTLF, the prediction process contains two main parts: recognition of interactive events and prediction of interactive trajectories. An integrated loss function is designed for multitask learning with the purpose of prediction and recognition. The proposed framework is verified using naturalistic driving data in the urban road. Experimental results show a superior performance of the GNN-MTLF compared to baseline methods and the potential for improving the road mobility.

Purpose To develop and evaluate a novel multitask deep learning framework for automated detection and localization of endoleaks at aortic digital subtraction angiography (DSA) performed during real-world endovascular aneurysm repair (EVAR) procedures for abdominal aortic aneurysm. Materials and Methods This retrospective study analyzed intraoperative aortic DSA images from patients undergoing EVAR (January 2017-December 2021). An expert panel assessed each sequence for endoleaks. Each sequence was processed into three input channels: peak density, time to peak, and area under the time-density curve, generating three two-dimensional perfusion maps per patient. These maps served as input into a convolutional neural network for binary detection (classification) and localization (regression) of endoleaks through multitask learning. Fivefold cross-validation was performed, with patients split 80:20 into training and testing datasets for each fold. Performance metrics included area under the receiver operating characteristic curve, F1 score, precision, and recall and were compared with human experts. Results The study included 220 patients (median age, 74 years [IQR, 68-79]; 181 male). Endoleaks were visible in 111 of 220 (50.5%) patients. The model identified and localized endoleaks with an area under the receiver operating characteristic curve of 0.85 ± 0.0031 (SD), F1 score of 0.78 ± 0.21, 95% precision, and 73% recall. Compared with the procedural team (94% precision, 63% recall), it had higher values in both metrics, with an F1 score within the human observer range (0.75-0.85). Balancing regression and classification by multitask learning delivered optimal results. The interobserver agreement among human experts was moderate (Fleiss κ = 0.404). Conclusion A novel, fully automated deep learning method accurately detected and localized endoleaks at DSA imaging from EVAR procedures. Keywords: Endoleaks, Deep Learning, Digital Subtraction Angiography, Endovascular Aneurysm Repair, Abdominal Aortic Aneurysm, Interventional Suite, Hybrid Operating Room Supplemental material is available for this article. © RSNA, 2025.

The human iris contains rich physiological information through its texture, color, and morphology. Traditional iris analysis methods are limited by computational inefficiency and poor scalability for multi-indicator prediction. This study aimed to develop and validate a lightweight, multi-task deep learning framework that can noninvasively predict multiple physiological indicators from high-resolution iris images. We developed a hybrid model combining ConvNeXt-Tiny and TinyViT backbones, trained on 30,353 iris images with a multi-task learning (MTL) strategy. To enhance performance with limited labeled data, semi-supervised learning with consistency regularization and pseudo-labeling was integrated. Fast Adaptive Multi-Task Optimization (FAMO) and Nash Multi-Task Learning (Nash-MTL) were used to balance task-specific losses. Feature vectors with 2400 dimensions — derived from texture, chromatic, and morphological cues — were compressed to 120 dimensions using Principal Component Analysis (PCA), preserving 96.2% of the data variance. This dimensionality reduction led to a 2.9-fold improvement in inference speed and reduced feature storage size by nearly 80% significantly enhancing computational efficiency without compromising predictive accuracy. The model achieved 89.1% average accuracy across 13 physiological tasks, with F1 scores of 0.91 for heart and 0.89 for brain indicators, outperforming fully supervised baselines. Among these, 7 organ-related indicators were validated using data from 60 Parkinson’s disease patients, showing strong left–right eye correlation ([Formula: see text], [Formula: see text]).