The Error Correcting Output Codes (ECOC) algorithm has played more and more important role in the artificial intelligence (AI) study and research fields in recent years due to its special properties and error correcting abilities. Various methods have been reported to improve its performance and classification accuracy. In this study, we developed a simplified ECOC encoding method to reduce the complexity in encoding process to reduce its random in initializing the coding matrix. Furthermore, we also suggested a fuzzy mapping decoding (FMD) method used for ECOC decoding process to improve its classification accuracy. By using this FMD, the classification accuracy can be improved up to 84% compared with the decoding method without using the FMD, which is 77%.

Sparse observations and coarse-resolution climate models limit effective regional decision-making, underscoring the need for robust downscaling. However, existing AI methods struggle with generalization across variables and geographies and are constrained by the quadratic complexity of Vision Transformer (ViT) self-attention. We introduce ORBIT-2, a scalable foundation model for global, hyper-resolution climate downscaling. ORBIT-2 incorporates two key innovations: (1) Residual Slim ViT (Reslim), a lightweight architecture with residual learning and Bayesian regularization for efficient, robust prediction; and (2) TILES, a tile-wise sequence scaling algorithm that reduces self-attention complexity from quadratic to linear, enabling long-sequence processing and massive parallelism. ORBIT-2 scales to 10 billion parameters across 65,536 GPUs, achieving up to 4.1 ExaFLOPS sustained throughput and 74–98% strong scaling efficiency. It supports downscaling to 0.9 km global resolution and processes sequences up to 4.2 billion tokens. On 7 km resolution benchmarks, ORBIT-2 achieves high accuracy with R2 scores in range of 0.98–0.99 against observation data.

Image retrieval (IR) for practical remote sensing (RS) should have high accuracy, storage, and calculation efficiency, while not relying on big annotations. However, current supervised and unsupervised RSIR methods do not yet fully meet these requirements. To this end, we propose a novel hashing-based IR approach via learning hash codes from open and representative self-supervised features. Specifically, we constructed a model out of a self-supervised pretrained backbone and a small multilayer perceptron (MLP)-based hashing learning neural network. Features from the frozen backbones were used to reconstruct a similarity matrix to guide the hash network learning. This way, the semantic structure can be preserved. To enhance the proposed approach, we propose the exploitation of global high-level semantic information within the similarity reconstruction process by introducing a small set of labeled datasets. Extensive comparative experiments on two commonly used RS image datasets demonstrate the outperformance of our proposed approach and its good balance between the retrieval accuracy and utilized annotations. In these two datasets, the labeled data required by our method accounts for less than 3% of that required by traditional methods, but our obtained mean average precision (mAP) can reach over 90%, which is close to that of current advanced supervised methods. In addition, we analyzed the specific effect of our design and the associated hyperparameters.

One of the unique traits of membrane proteins is that a significant fraction of their hydrophobic amino acids is exposed to the hydrophobic core of lipid bilayers rather than being embedded in the protein interior, which is often not explicitly considered in the protein structure and function predictions. Here, we propose a characteristic and predictive quantity, the membrane contact probability (MCP), to describe the likelihood of the amino acids of a given sequence being in direct contact with the acyl chains of lipid molecules. We show that MCP is complementary to solvent accessibility in characterizing the outer surface of membrane proteins, and it can be predicted for any given sequence with a machine learning-based method by utilizing a training dataset extracted from MemProtMD, a database generated from molecular dynamics simulations for the membrane proteins with a known structure. As the first of many potential applications, we demonstrate that MCP can be used to systematically improve the prediction precision of the protein contact maps and structures.

Thermomechanical processes (TMPs) such as resistance spot welding (RSW) and hot stamping are widely used in automotive manufacturing. Recent advancement in sensing technology has led to an increasing adoption of thermographic cameras to capture the infrared (IR) radiation of a metal part (or component of a part) during its thermomechanical processing or immediately after the process when the part is still hot. Detecting the object(s) of interest from raw IR images is an essential step in analyzing these data. Deep learning (DL) has been a recent success for object detection (OD), but the application of DL-based OD for industrial IR images in manufacturing is largely lagging behind. The major contribution of this work, which is also the distinction from previous OD studies, is the capability of building the OD model with unlabeled IR images, i.e., imaging data without accurate information indicating the object position. The architecture of Unsupervised IR Image Net (UIR-Net) is designed to accommodate the unique characteristics of IR images from TMPs in manufacturing. This study presents a novel method for OD in unlabeled IR images from TMPs. The proposed method, called UIR-Net, consists of two components: label generation and DL model construction. Two case studies from automotive manufacturing, RSW and hot stamping, are reported to demonstrate the feasibility and effectiveness of the proposed method. Note to Practitioners—This article was motivated by the problem of detecting objects such as weld nugget or metal piece in infrared (IR) imaging of thermomechanical processes (TMPs) in automotive manufacturing. The method is applicable to in situ IR images or videos that contain one or more objects to be detected. It only requires that the data are in image form and come from TMPs. Currently, there is no existing deep learning (DL)-based method for generic object detection (OD) in unlabeled IR images from TMPs. The proposed method takes advantages of the recent advancement in DL. This article suggests a systematic approach to build a DL-based OD model, named Unsupervised IR Image Net (UIR-Net), to extract objects from raw IR images collected for TMPs. A step-by-step procedure is given in this article to guide users through label generation, data quality evaluation, and model training to establish the proposed UIR-Net model. Results from resistance spot welding and hot stamping suggest that this approach is feasible and effective. It is one of the few generic OD works designed for manufacturing applications. Simple implementation, feasibility, and effectiveness make this method a suitable candidate for online data analytics and process monitoring in a wide range of manufacturing applications.

Autonomous industrial assembly relies on the precise measurement of spatial constraints as designed by computer-aided design (CAD) software such as SolidWorks. This paper proposes a framework for an intelligent industrial robot to understand the spatial constraints for model assembly. An extended generative adversary network (GAN) with a 3D long short-term memory (LSTM) network was designed to composite 3D point clouds from a single RGB-D scan. The spatial constraints of the segmented point clouds are identified by a neural-logic network that incorporates general knowledge of spatial constraints in terms of first-order logic. The model was designed to comprehend a complete set of spatial constraints that are consistent with industrial CAD software, including left, right, above, below, front, behind, parallel, perpendicular, concentric, and coincident relations. The accuracy of 3D model composition and spatial constraint identification was evaluated by the RGB-D scans and 3D models in the ABC dataset. The proposed model achieved 57.23% intersection over union (IoU) in 3D model composition, and over 99% in comprehending all spatial constraints.

BACKGROUND: Mutations in an enzyme target are one of the most common mechanisms whereby antibiotic resistance arises. Identification of the resistance mutations in bacteria is essential for understanding the structural basis of antibiotic resistance and design of new drugs. However, the traditionally used experimental approaches to identify resistance mutations were usually labor-intensive and costly. RESULTS: We present a machine learning (ML)-based classifier for predicting rifampicin (Rif) resistance mutations in bacterial RNA Polymerase subunit β (RpoB). A total of 186 mutations were gathered from the literature for developing the classifier, using 80% of the data as the training set and the rest as the test set. The features of the mutated RpoB and their binding energies with Rif were calculated through computational methods, and used as the mutation attributes for modeling. Classifiers based on five ML algorithms, i.e. decision tree, k nearest neighbors, naïve Bayes, probabilistic neural network and support vector machine, were first built, and a majority consensus (MC) approach was then used to obtain a new classifier based on the classifications of the five individual ML algorithms. The MC classifier comprehensively improved the predictive performance, with accuracy, F-measure and AUC of 0.78, 0.83 and 0.81for training set whilst 0.84, 0.87 and 0.83 for test set, respectively. CONCLUSION: The MC classifier provides an alternative methodology for rapid identification of resistance mutations in bacteria, which may help with early detection of antibiotic resistance and new drug discovery.

Semantic segmentation of images is an important computer vision task that emerges in a variety of application domains such as medical imaging, robotic vision and autonomous vehicles to name a few. While these domain-specific image analysis tasks involve relatively small image sizes (∼ 102 × 102), there are many applications that need to train machine learning models on image data with extents that are orders of magnitude larger (∼ 104 × 104). Training deep neural network (DNN) models on large extent images is extremely memory-intensive and often exceeds the memory limitations of a single graphical processing unit, a hardware accelerator of choice for computer vision workloads. Here, an efficient, sample parallel approach to train U-Net models on large extent image data sets is presented. Its advantages and limitations are analyzed and near-linear strong-scaling speedup demonstrated on 256 nodes (1536 GPUs) of the Summit supercomputer. Using a single node of the Summit supercomputer, an early evaluation of a recently released model parallel framework called GPipe is demonstrated to deliver ∼ 2X speedup in executing a U-Net model with an order of magnitude larger number of trainable parameters than reported before. Performance bottlenecks for pipelined training of U-Net models are identified and mitigation strategies to improve the speedups are discussed. Together, these results open up the possibility of combining both approaches into a unified scalable pipelined and data parallel algorithm to efficiently train U-Net models with very large receptive fields on data sets of ultra-large extent images.

Understanding spatial relations of objects is critical in many robotic applications such as grasping, manipulation, and obstacle avoidance. Humans can simply reason object's spatial relations from a glimpse of a scene based on prior knowledge of spatial constraints. The proposed method enables a robot to comprehend spatial relationships among objects from RGB-D data. This paper proposed a neural-logic learning framework to learn and reason spatial relations from raw data by following logic rules on spatial constraints. The neural-logic network consists of three blocks: grounding block, spatial logic block, and inference block. The grounding block extracts high-level features from the raw sensory data. The spatial logic blocks can predicate fundamental spatial relations by training a neural network with spatial constraints. The inference block can infer complex spatial relations based on the predicated fundamental spatial relations. Simulations and robotic experiments evaluated the performance of the proposed method.

Deep learning has great potential to reshape manufacturing industries. In this article, we present DeepWelding, a novel framework that applies deep learning techniques to improve gas tungsten arc welding process monitoring and penetration detection using multisource sensing images. The framework is capable of analyzing multiple types of optical sensing images synchronously and consists of three deep learning enhanced consecutive phases: image preprocessing, image selection, and weld penetration classification. Specifically, we adopted generative adversarial networks (pix2pix) for image denoising and classic convolutional neural networks (AlexNet) for image selection. Both pix2pix and AlexNet delivered satisfactory performance. However, five individual neural networks with heterogeneous architectures demonstrated inconsistent generalization capabilities in the classification phase when holding out multisource images generated with specific experimental settings. Therefore, two ensemble methods combining multiple neural networks are designed to improve the model performance on unseen data collected from different experimental settings. We have also found that the quality of model prediction is heavily influenced by the data stream collection environment. We think these findings are beneficial for the broad intelligent welding community.

A comparison study among type 1 fuzzy interpolations (T1FI), interval type 2 fuzzy interpolations (IT2FI) and neural network fitting functions (NNFF) applied on modeless robots calibrations is proposed. Traditional robots calibration implements either model or modeless method. The compensation of position error in modeless method is to move the robot's end-effector to a target position in the robot workspace, and to find the target position error based on the measured neighboring 4-grid-point errors around the target position. A camera or other measurement device is attached on the robot's end-effector to find and measure the neighboring position errors, and compensate the target position with various error interpolation methods. By using the NNFF technique provided in this paper, the accuracy of the position error compensation can be greatly improved, which has been confirmed by the simulation results given in this paper. Compared with some other popular interpolation methods, this NNFF technique is a better choice. The simulation results show that more accurate compensation result can be achieved using this technique compared with the type-1 fuzzy and interval type 2 fuzzy interpolation methods.

Motivation: Cell movement in the early phase of Caenorhabditis elegans development is regulated by a highly complex process in which a set of rules and connections are formulated at distinct scales. Previous efforts have demonstrated that agent-based, multi-scale modeling systems can integrate physical and biological rules and provide new avenues to study developmental systems. However, the application of these systems to model cell movement is still challenging and requires a comprehensive understanding of regulatory networks at the right scales. Recent developments in deep learning and reinforcement learning provide an unprecedented opportunity to explore cell movement using 3D time-lapse microscopy images. Results: We present a deep reinforcement learning approach within an agent-based modeling system to characterize cell movement in the embryonic development of C.elegans. Our modeling system captures the complexity of cell movement patterns in the embryo and overcomes the local optimization problem encountered by traditional rule-based, agent-based modeling that uses greedy algorithms. We tested our model with two real developmental processes: the anterior movement of the Cpaaa cell via intercalation and the rearrangement of the superficial left-right asymmetry. In the first case, the model results suggested that Cpaaa's intercalation is an active directional cell movement caused by the continuous effects from a longer distance (farther than the length of two adjacent cells), as opposed to a passive movement caused by neighbor cell movements. In the second case, a leader-follower mechanism well explained the collective cell movement pattern in the asymmetry rearrangement. These results showed that our approach to introduce deep reinforcement learning into agent-based modeling can test regulatory mechanisms by exploring cell migration paths in a reverse engineering perspective. This model opens new doors to explore the large datasets generated by live imaging. Availability and implementation: Source code is available at https://github.com/zwang84/drl4cellmovement. Supplementary information: Supplementary data are available at Bioinformatics online.

This paper provides an effective method to evaluate the input quantization and multiplication roundoff errors for two-dimensional (2-D) separable-in-denominator recursive digital (SDRD) filters. This work is based on the partial fraction expansion of 2-D SDRD filters [1] and their modular implementation structures [2] proposed by the authors. Equations are derived for the calculation of roundoff errors for various implementations of 2-D SDRD filters. An example is given to demonstrate the application of the proposed method.

It is critically important to develop intelligent mobile robots to effectively reduce the effects of nuclear radiations to perform desired rescuing tasks in critical emergency situations. One of the good examples is the recent nuclear disposing damage to the Japanese nuclear power plants after the earth quick occurred in the northeast of Japan. In this study, a nested fuzzy logic control system is designed for mobile robots to effectively reduce the effects of those lower-frequency nuclear radiations on rescuing robots. A coarse and a fine fuzzy controller are combined together to realize this objective. The simulation and experimental results of this study confirmed the effectiveness of this nested fuzzy control system in reduction of the effects of lower-frequency nuclear radiations on mobile robots. This is the first paper related to this topic to be presented.

Complex spatial control problems can be computationally intensive. Timely response in urgent spatial control situations such as wildfire control poses great challenges for the efficient solving of spatial control problems. Web-based and service-oriented architectures of integrating geographic information system (GIS) clients and parallel computing resources have been suggested as an effective paradigm to solve computationally intensive spatial problems. Such real-time coupling framework is highly dependent upon interactivity and on-demand availability of dedicated parallel computing resources appropriate for the problem. We present an approach to enhancing the efficiency of solving spatial control problems while offering another coupling framework of integrating computing resources from desktop GIS and parallel computing environments to alleviate such dependency. Specifically, a model knowledge database is developed to bridge the gap between desktop GIS models and parallel computing resources. Desktop GIS models can iteratively improve themselves by steering rules retrieved from the model knowledge database. To examine its effectiveness, we applied the framework to a wildfire control case. Simulation results show dramatic reduction in computation time of the improved desktop GIS model, and indicate that desktop GIS models enhanced by model knowledge databases can be useful in providing timely assistance on computationally intensive spatial control problems.

The weapon identification and selection issue is an important and strategic component and has a significant impact on the efficiency of defense system in US. The main purpose of this research is to develop a universal model and system to effectively assess, evaluate and identify the optimal weapons from a large collection of available weapon systems that have multiple criteria based on a fuzzy multiple criteria decision making (FMCDM) model. A simple but effective weight estimation method is adopted in this paper to make this selection more objective and reliable. There are some different weight estimation methods reported by researchers, however, most methods need a lot of mathematical operations and make the process very time consuming and even more complicated. In this paper, we utilized a weight estimation method based on the paired comparison matrix to simplify this estimation process. The evaluation and selection process can be significantly simplified and improved by using this method.

The paper provides a comparison between a novel technique used for the pose error measurements and compensations of robots based on a fuzzy error interpolation method and some other popular interpolation methods. By using the proposed fuzzy error interpolation technique, the accuracy of the pose-error compensation can be improved, which has been confirmed by the simulation results given in this paper, compared with other interpolation methods. A comparison study among various interpolation methods, such as trilinear, cubic spline, and the fuzzy error interpolation technique is also made and discussed via simulation. The simulation results show that more accurate measurement and compensation results can be achieved using the fuzzy error interpolation technique compared with its trilinear and cubic spline counterparts.

In this paper, a number of algorithmic options for fuzzy logic control applications are presented. The emphasis is on the analyses of the computational load and memory requirements of all the processing stages of fuzzy logic control algorithms. Comparisons are made among commonly used techniques and recommendations are provided. The results could be used to guide the selection of a fuzzy logic algorithm for a specific application.

This paper provides a comparison between a novel technique used for the pose-error measurements and compensations of robots based on a fuzzy-error interpolation method and some other popular interpolation methods. A traditional robot calibration implements either model or modeless methods. The measurement and compensation of pose errors in a modeless method moves the robot's end-effector to the target poses in the robot workspace and measures the target position and orientation errors using some interpolation techniques in terms of the premeasured neighboring pose errors around the target pose. For the measurement purpose, a stereo camera or other measurement devices, such as a coordinate-measurement machine (CMM) or a laser-tracking system (LTS), can be used to measure the pose errors of the robot's end-effector at predened grid points on a cubic lattice. By the use of the proposed fuzzy-error interpolation technique, the accuracy of the pose-error compensation can be improved in comparison with other interpolation methods, which is conrmed by the simulation results given in this paper. A comparison study among most popular interpolation methods used in modeless robot calibrations, such as trilinear, cubic spline, and the fuzzy-error interpolation technique, is also made and discussed via simulations. The simulation results show that more accurate measurement and compensation results can be achieved using the fuzzy-error interpolation technique compared with its trilinear and cubic-spline counterparts.

In collaboration with Norfolk State University (NSU) and University of Kentucky (UKY), SURA is leading a short-term pilot project to establish a distributed Microsoft high performance computing (HPC) test bed in support of biomedical and geosciences research. The goals of this pilot are to (1) establish a test bed using Microsoft HPC2008, (2) investigate the integration of Microsoft with other HPC environments, and (3) demonstrate the utility of this environment for supporting academic research. Through the integration of the Microsoft environment with the Globus-based SURAgrid, we hope to broaden the pool of HPC resources available to SURA researchers and to broaden engagement.

Traditional robot calibration implements model and modeless methods. The compensation of position error in modeless method is to move the end-effector of robot to the target position in the workspace, and to find the position error of that target position by using a bilinear interpolation method based on the neighboring 4-point's errors around the target position. A camera or other measurement devices can be utilized to find or measure this position error, and compensate this error with the interpolation result. This paper provides a novel fuzzy interpolation method to improve the compensation accuracy obtained by using a bilinear interpolation method. A dynamic online fuzzy inference system is implemented to meet the needs of fast real-time control system and calibration environment. The simulated results show that the compensation accuracy can be greatly improved by using this fuzzy interpolation method compared with the bilinear interpolation method.

DNA microarray technology enables a parallel analysis of the expression of genes in an organism. The wealth of spatio-temporal data generated by this technology allows researchers to potentially reverse engineer the genetic network. Fuzzy logic has been proposed as a method to analyze the relationships between genes. This method can identify interacting genes that fit a known fuzzy model of gene interaction by testing all combinations of gene expression profiles. However, this approach is slow and computationally complex. This paper introduces improvements made in terms of reducing computation time and generalizing the gene regulatory model to accommodate co-activators and co-repressors. Improvement in computation time is achieved by using clustering as a pre-processing method, thereby reducing the total number of gene combinations analyzed. This will allow the algorithm to run in a shorter amount of time with minimal effect on the results. The proposed technique will pave the way towards the creation of a generalized gene interaction model that can accommodate any combination of genes.

This paper presents a novel clustering technique known as adaptive double self-organizing map (ADSOM). ADSOM has a flexible topology and perform clustering and cluster visualization simultaneously, thereby requiring no a priori knowledge about the number of clusters. ADSOM combines features of the popular self-organizing map (SOM) with two-dimensional position vectors, which serve as a visualization tool to accurately determine the number of clusters present in the data. ADSOM updates its free parameters during training and it allows convergence of its position vectors to a fairly consistent number of clusters provided that its initial number of nodes is greater than the expected number of clusters. A novel index is introduced based on hierarchical clustering of the final locations of position vectors. The index allows automatic detection of the number of clusters, thereby reducing human error that could be incurred from counting cluster visually. The reliance of ADSOM in identifying the number of clusters is proven by applying it to publicly available yeast gene expression data.

This paper describes a novel technique for the position error compensations of the robot and manipulator calibration process based on a fuzzy error interpolation method. Traditional robots calibration implements model and modeless methods. Bilinear and cubic splines are the two popular techniques for the compensation of position error in modeless method. By using the fuzzy error interpolation technique provided in this paper, the accuracy of the position error compensation can be greatly improved. The comparison between the two popular traditional interpolation methods and this fuzzy error interpolation technique is made via simulation for three commonly used error models. The simulation results show that more accurate compensation result can be achieved using the fuzzy error interpolation technique compared with the bilinear and cubic spline interpolation methods.