41 pubs · 276 cites · h-index 9
A critical bottleneck for scientific progress is the costly nature of computer simulations for complex systems. Surrogate models provide an appealing solution: such models are trained on simulator evaluations, then used to emulate and quantify uncertainty on the expensive simulator at unexplored inputs. In many applications, one often has available data on related systems. For example, in designing a new jet turbine, there may be existing studies on turbines with similar configurations. A key question is how information from such “source” systems can be transferred for effective surrogate training on the “target” system of interest. We thus propose a new LOcal transfer Learning Gaussian Process (LOL-GP) model, which leverages a carefully designed Gaussian process to transfer such information for surrogate modeling. The key novelty of the LOL-GP is a latent regularization model, which identifies regions where transfer should be performed and regions where it should be avoided. Such a “local transfer” property is present in many scientific systems: for certain parameters, systems may behave similarly and thus transfer is beneficial; for other parameters, they may behave differently and thus transfer is detrimental. By accounting for local transfer, the LOL-GP can temper the risk of “negative transfer,” i.e., the risk of worsening predictive performance from information transfer. We derive a Gibbs sampling algorithm for efficient posterior predictive sampling on the LOL-GP for both the multisource and multifidelity transfer settings. We then show, via a suite of numerical experiments and an application for jet turbine design, the improved surrogate performance of the LOL-GP over existing methods.
The YOLO-LF model enhances small target detection, indicating its value for agricultural disease management.
Road crack detection is crucial for ensuring pavement safety and optimizing maintenance strategies. This study investigated the impact of image preprocessing methods and dataset balance on the performance of YOLOv8s-based crack detection. Four datasets (CFD, Crack500, CrackTree200, and CrackVariety) were evaluated using three image formats: RGB, grayscale (five conversion methods), and binarized images. The experimental results indicate that RGB images consistently achieved the highest detection accuracy, confirming that preserving color-based contrast and texture information benefits YOLOv8's feature extraction. Grayscale conversion showed dataset-dependent variations, with different methods performing best on different datasets, while binarization generally degraded detection accuracy, except in the balanced CrackVariety dataset. Furthermore, this study highlights that dataset balance significantly impacts model performance, as imbalanced datasets (CFD, Crack500, CrackTree200) led to biased predictions favoring dominant crack classes. In contrast, CrackVariety's balanced distribution resulted in more stable and generalized detection. These findings suggest that dataset balance has a greater influence on detection accuracy than preprocessing methods. Future research should focus on data augmentation and resampling strategies to mitigate class imbalance, as well as explore multi-modal fusion approaches for further performance enhancements.
Plant leaf disease detection plays a crucial role in agricultural production and food safety. With the rapid development of the Internet of Things (IoT) and advancements in sensor technology, plant leaf disease detection has become capable of remote and real-time monitoring, greatly enhancing the efficiency and precision of agricultural production. However, current detection methods face issues of insufficient accuracy and high model complexity, limiting their effectiveness and scalability in practical applications. To address these challenges, this paper proposes a novel network architecture, LeafMamba, which integrates the core ideas of the Mamba structure and overcomes the limitations of Transformer in complex scenarios. Through the innovatively designed LF-Mamba structure, LeafMamba efficiently captures multi-scale information and progressively enhances the spatial resolution of feature maps in the decoder, thereby improving the model’s ability to extract complex disease features. Experimental results on the Plant Pathology 2020 - FGVC7 and Plant Pathology 2021 - FGVC8 datasets show that LeafMamba achieves performance of 89.5% and 92.5% in mAP@0.5, respectively, significantly outperforming existing methods. Furthermore, the model reduces the number of parameters substantially, demonstrating higher computational efficiency.
Multioutput Gaussian process (MGP) is commonly used as a transfer learning method to leverage information among multiple outputs. A key advantage of MGP is providing uncertainty quantification for prediction, which is highly important for subsequent decision-making tasks. However, traditional MGP may not be sufficiently flexible to handle multivariate data with dynamic characteristics, particularly when dealing with complex temporal correlations. Additionally, because some outputs may lack correlation, transferring information among them may lead to negative transfer. To address these issues, this study proposes a nonstationary MGP model that can capture both the dynamic and sparse correlation among outputs. Specifically, the covariance functions of MGP are constructed using convolutions of time-varying kernel functions. Then a dynamic spike-and-slab prior is placed on correlation parameters to automatically decide which sources are informative to the target output in the training process. An expectation-maximization (EM) algorithm is proposed for efficient model fitting. Both numerical studies and real cases demonstrate its efficacy in capturing dynamic and sparse correlation structure and mitigating negative transfer for high-dimensional time-series data modeling. A mountain-car reinforcement learning case highlights that transferring knowledge from source expertise can enhance the efficiency of the target decision-making process. Our method holds promise for application in more complex multitask decision-making challenges within nonstationary environments in the future. History: Rema Padman served as the senior editor for this article. Funding: This work was supported by NSFC [Grants NSFC-72171003, NSFC-71932006, and NSFC-72101147]. Data Ethics & Reproducibility Note: The code capsule is available on Code Ocean at https://doi.org/10.24433/CO.4010696.v1 and in the e-Companion to this article (available at https://doi.org/10.1287/ijds.2023.0022 ).
In this study, we propose a safety-critical compliant control strategy designed to strictly enforce interaction force constraints during the physical interaction of robots with unknown environments. The interaction force constraint is interpreted as a new force-constrained control barrier function (FC-CBF) by exploiting the generalized contact model and the prior information of the environment, i.e., the prior stiffness and rest position, for robot kinematics. The difference between the real environment and the generalized contact model is approximated by constructing a tracking differentiator, and its estimation error is quantified based on Lyapunov theory. By interpreting strict interaction safety specification as a dynamic constraint, restricting the desired joint angular rates in kinematics, the proposed approach modifies nominal compliant controllers using quadratic programming, ensuring adherence to interaction force constraints in unknown environments. The strict force constraint and the stability of the closed-loop system are rigorously analyzed. Experimental tests using a UR3e industrial robot with different environments verify the effectiveness of the proposed method in achieving the force constraints in unknown environments.
Review demonstrates enhanced refractory performance in fiber concrete mixes, highlighting the critical influence of fiber type and mix ratios on structural fire resistance.
In this paper, a safety-critical compliance control with force constraint is proposed for the interaction task of a manipulator in the Cartesian space. Different from the conventional impedance control approach, the interaction force constraint is considered and interpreted by designing a new Force Constrained Control Barrier Function (FCCBF) with exploiting the environment model. By formulating and solving a quadratic programming with modifying the baseline impedance controller under the FCCBF constraint, the proposed controller can not only make manipulator's interaction process satisfy expected compliance without exceeding permissible contact force range, but can also improve the dynamic characteristics of the system by eliminating force overshooting phenomena. Furthermore, with using the technique of nonlinear disturbance observer, a force sensor-less FCCBF approach is also provided, where the contact force is treated as an external disturbance. Finally, the effectiveness of the proposed algorithms is verified by both simulations and experimental tests on a 6 degree-of-freedom manipulator.
Methodological study demonstrates improved prediction accuracy in multi-output systems via regularized convolution processes, suggesting reliable transfer learning across inconsistent domains.
Porosity is one of the most critical quality issues in Additive Manufacturing (AM). As process parameters are closely related to porosity formation, it is vitally important to study their relationship for better process optimization. In this article, motivated by the emerging application of metal AM, a three-level hierarchical mixed-effects modeling approach is proposed to characterize the relationship between microstructural images and process parameters for porosity prediction and microstructure reconstruction. Specifically, a Two-Point Correlation Function (TPCF) is used to capture the morphology of the pores quantitatively. Then, the relationship between the TPCF profile and process parameters is established. A blocked Gibbs sampling approach is developed for parameter inference. Our modeling framework can reconstruct the microstructure based on the predicted TPCF through a simulated annealing optimization algorithm. The effectiveness and advantageous features of our method are demonstrated by both the simulation study and the case study with real-world data from metal AM applications.
Multi-tenancy is one of the most important features in Software as a Service(SaaS). In order to reduce cost, SaaS service vendors typically implement multi-tenant enabled database on top of conventional SQL database, and Universal Table is one of the most commonly used implementation technique. The key idea is to map all tenant data into one single table. At the application level, such a scheme-mapping is implemented by translating SQL statements. Previously, researchers have introduce methods to the translation of SQL query. In this work, we extend previous work by proposing a method that support the translation of full SQL syntax, including both data manipulation language (DML) and data definition language (DDL). Our method is based on syntax-directed translation on the context-free grammar of full SQL syntax. We have implemented our method as an automatic translation tool based on the ANTLR framework. In order to evaluate the effectiveness and efficiency of our method, we conduct an empirical study using the TPC-C benchmark. The results show that our method is scalable to the number of tenants and the size of database.
The widespread recognition of the smart contracts has established their importance in the landscape of next generation blockchain technology. However, writing a correct smart contract is notoriously difficult. Moreover, once a state-changing transaction is confirmed by the network, the result is immutable. For this reason, it is crucial to perform a thorough testing of a smart contract application before its deployment. This paper's focus is on the test coverage criteria for smart contracts, which are objective rules that measure test quality. We analyze the unique characteristics of the Ethereum smart contract program model as compared to the conventional program model. To capture essential control flow behaviors of smart contracts, we propose the notions of whole transaction basis path set and bounded transaction interaction. The former is a limited set of linearly independent inter-procedural paths from which the potentially infinite paths of Ethereum transactions can be constructed by linear combination, while the latter is the permutations of transactions within a certain bound. Based on these two notions, we define a family of path-based test coverage criteria. Algorithms are given to the generation of coverage requirements. A case study is conducted to compare the effectiveness of the proposed test coverage criteria with random testing and statement coverage testing.
Lines are significant features enclosing high-level information in an image. The line segment Detector (LSD) Algorithm with low error rate is a widely used method to extract lines in images effectively and accurately. However, the algorithm on PC performs too costly both in time and resources for the real-time video processing. This paper provides a fast and resource-efficient hardware implementation solution for a modified LSD algorithm on Field Programmable Gate Arrays (FPGA) for real-time line detection. The task-level pipeline structures are exploited fully in a stream process mapped to the hardware architecture free of frame buffer. Our proposed hardware implementation processes in a stream-in-stream-out manner with little consumption of the on-chip block RAM to store intermediate values. We first employ hardware Gaussian filter and adjust Canny edge detection to obtain an edge map at single-pixel width. Then, a novel structure of region growing model based on dynamic rooted tree is used to detect line segment regions accurately with a latency of only a few rows of pixels. The low cost in time, on-chip resources, and power consumption makes our proposed algorithm suitable for portable real-time streaming video processing applications using line segment features, such as Lane departure warning systems. It can also be applied in real-time machine vision systems that use line segments information for further recognition or stereo correspondence and many others. The proposed algorithm is synthesized and tested on XC7Z020 FPGA with high reliability, accuracy speed, and low cost in both resources and energy.
Wireless sensors and mobile devices have been widely deployed as data collecting devices for monitoring real world systems. A large amount of stream data is generated in real-time, which has to be processed in real-time as well. One of the common processing operations is clustering that automatically groups the elements of a stream into a number of clusters in general. Elements of the same cluster have maximum similarity and elements of different clusters have minimum similarity. This paper proposes an on-demand framework (SRAStream) based on the concept drifting detection mechanism. The concept drifting detection algorithm is used to measure the distance of the new clusters for the current data and that of the existing clusters. Only when a concept drifting occurs will the re-clustering be performed to identify new clusters. SRAStream thus avoids the unnecessary computation intensive re-clustering calculation. Experiments suggest that the proposed framework does work well and improve the processing speed greatly in data streams clustering.
Thermal elasticity of two kinds of bonding-type glassy alloys, Zr55Cu30Al10Ni5 and Pd40Cu30Ni10P20, during progressive devitrification was evaluated by Poisson’s ratio, Debye temperature, and internal frictions in terms of topological change in metallic clusters. Both alloys showed opposite behavior for thermal elasticity: thermal softening associated with relaxation of icosahedral clusters and thermal-induced entropy elasticity associated with the rotational and vibrational motions of polyhedral clusters, and an increment of internal friction based on electron-longitudinal phonon interaction derived from the variation amount of potential energy among atoms, showing vanishment of free electrons.