Retrieval-Augmented Generation (RAG) is a promising technique for mitigating two key limitations of large language models (LLMs): outdated information and hallucinations. RAG system stores documents as embedding vectors in a database. Given a query, search is executed to find the most related documents. Then, the topmost matching documents are inserted into LLMs’ prompt to generate a response. Efficient and accurate searching is critical for RAG to get relevant information. We propose a cost-effective searching algorithm for retrieval process. Our progressive searching algorithm incrementally refines the candidate set through a hierarchy of searches, starting from low-dimensional embeddings and progressing into a higher, target-dimensionality. This multi-stage approach reduces retrieval time while preserving the desired accuracy. Our findings demonstrate that progressive search in RAG systems achieves a balance between dimensionality, speed, and accuracy, enabling scalable and high-performance retrieval even for large databases. Our code is available at https://github.com/taeheej/Progressive-searching-for-Retrieval-in-RAG.

Graph neural networks (GNNs) have received increasing interest in the medical imaging field given their powerful graph embedding ability to characterize the non-Euclidean structure of brain networks based on magnetic resonance imaging (MRI) data. However, previous studies are largely node-centralized and ignore edge features for graph classification tasks, resulting in moderate performance of graph classification accuracy. Moreover, the generalizability of GNN model is still far from satisfactory in brain disorder [e.g., autism spectrum disorder (ASD)] identification due to considerable individual differences in symptoms among patients as well as data heterogeneity among different sites. In order to address the above limitations, this study proposes a novel adversarial learning-based node-edge graph attention network (AL-NEGAT) for ASD identification based on multimodal MRI data. First, both node and edge features are modeled based on structural and functional MRI data to leverage complementary brain information and preserved in the constructed weighted adjacent matrix for individuals through the attention mechanism in the proposed NEGAT. Second, two AL methods are employed to improve the generalizability of NEGAT. Finally, a gradient-based saliency map strategy is utilized for model interpretation to identify important brain regions and connections contributing to the classification. Experimental results based on the public Autism Brain Imaging Data Exchange I (ABIDE I) data demonstrate that the proposed framework achieves a classification accuracy of 74.7% between ASD and typical developing (TD) groups based on 1007 subjects across 17 different sites and outperforms the state-of-the-art methods, indicating satisfying classification ability and generalizability of the proposed AL-NEGAT model. Our work provides a powerful tool for brain disorder identification.

Due to less degradation data and the inconsistent data distribution of different bearings, remaining useful life (RUL) prediction methods based on deep learning still do not yield satisfactory predictive results. Using RUL prediction model trained with one bearing sample but tested with another bearing sample is challenging. To solve this problem, in this paper a new deep transfer learning-based RUL prediction method (DTL-RULPM) is proposed. We adopt min-max normalization to normalize the original vibration data of bearing. A three-layer sparse autoencoder is designed to extract the deep features of the source domain. Random data with standard normal distribution is generated with the consistent dimension of the high-dimensional features of the source domain. Maximum mean discrepancy (MMD) is used to minimize the probability distribution distance between the features of the source domain and the randomly generated data, such that the model can learn domain-invariant features of different bearings. Then we adopt a bi-directional long and short-term memory (Bi-LSTM) network to predict the RUL of the bearing. We use the IEEE PHM Challenge 2012 dataset to verify the proposed method. The results demonstrate that the proposed method improves the RUL prediction accuracy and robustness of different bearings.

In recent years, transfer learning has been widely used in cross-domain fault diagnosis to solve the problem of insufficient training data. Existing studies focus on the homogeneous transfer fault diagnosis of the same component with different operating conditions. However, when the source and target domain are from two different components, the feature space and category space of the two domains are completely different, which causes the challenging problem of unsupervised heterogeneous transfer fault diagnosis. We propose a graph Laplacian common subspace based unsupervised heterogeneous transfer learning model (GL-HTLM). Firstly, pseudo-labels are designed for the unlabeled samples of target domain using the Gaussian mixture model to learn the distribution characteristics of the original vibration signals. Secondly, a deep convolutional neural network is designed to extract the high-dimensional features of the labeled samples of source domain and the pseudo-labeled samples of target domain. Finally, a common latent attributes space (CLAS) is generated through near-binary feature representation learning to extract the latent attributes of the source and target domain. According to the similarity of any two samples in CLAS, we further define graph Laplacian loss to maximize the inter-category distances while minimizing the intra-category distances. Therefore, the two domains with different category spaces are strongly consistent in the CLAS, so as to classify samples of two domains. In order to validate the proposed method, four heterogeneous transfer fault diagnosis experiments are carried out using bearing dataset and gear dataset. Results demonstrate that our proposed model is superior to existing methods.

The compound fault signal of bearings is coupled and complex, thereby compound fault diagnosis is a difficult problem in bearing fault diagnosis. The existing deep learning models can extract fault features when there are a large number of labeled compound fault samples. In the industrial scenarios, collecting and labeling sufficient compound fault samples are unpractical. Using the model trained on single fault sample to identify unknown compound fault is challenging and innovative. To address this problem, we propose a Zero-shot Learning Compound Fault Diagnosis Model of bearing (ZLCFDM). First, we design a semantic encoding method to express the semantic vectors of single fault and compound fault according to the fault characteristics. Second, a convolutional neural network is designed to extract the time-frequency visual features of compound fault signal. Then we embed the semantic vector of the fault into the visual space of the fault data. The cosine distance is merged into K-nearest neighbor (KNN) to measure the distance between the visual features and the semantic vectors of the compound faults, such that the model can identify the categories of unknown compound faults. To validate the proposed method, we conduct experiments on self-built testbed. The results demonstrate that the identification accuracy of compound fault can reach 77.73% when the model trained without any compound fault samples. This is the first time to propose the compound fault diagnosis of bearing base on zero-shot learning.

Abstract Edge computing perfectly integrates cloud computing centers and edge-end devices together, but there are not many related researches on how the edge-end node devices work to form an edge network and what the protocols used to implement the communication among nodes in the edge network. Aiming at the problem of coordinated communication among edge nodes in the current edge computing network architecture, this paper proposes an edge network routing and forwarding protocol based on target tracking scenarios. This protocol can meet the dynamic changes of node locations, and the elastic expansion of node scale. Individual node failures will not affect the overall network, and the network ensures efficient real-time with less communication overhead. The experimental results display that the protocol can effectively reduce the communications volume of the edge network, improve the overall efficiency of the network, and set the optimal sampling period, so as to ensure that the network delay is minimized.

Swarming of unmanned aerial vehicles (UAVs) has become a popular topic for its practicability in many scenarios. However an easy-to-use yet powerful swarm testbed is not widely available for most of the research labs globally. In this paper, a UAV swarm testbed is presented, which consist of an ultra-wide-band (UWB) positioning system, multiple UAVs and finite state machine based on Stateflow/Simulink. There are three important features that make this system standing out: the one is the tracking accuracy is the best as far as we know in the UWB positioning category; the second is that the system can resist magnetometer interference, i.e. the yaw angle is not effected by magnetic interferences.; the last one is that the finite state machine is ready for the general swarm experiments. The system is easy to setup with auto calibration, not limited by the size of the testing space, and can be used for both indoor and outdoor. It is believed that the proposed testbed can be adopted by swarm research labs in different scenarios in multiple scales.

Distributed and parallel computing is becoming more important with the availability of extremely large data sets. In this article, we consider this problem for high-dimensional linear quantile regression. We work under the assumption that the coefficients in the regression model are sparse; therefore, a LASSO penalty is naturally used for estimation. We first extend the debiasing procedure, which is previously proposed for smooth parametric regression models to quantile regression. The technical challenges include dealing with the nondifferentiability of the loss function and the estimation of the unknown conditional density. In this article, the main objective is to derive a divide-and-conquer estimation approach using the debiased estimator which is useful under the big data setting. The effectiveness of distributed estimation is demonstrated using some numerical examples.

Unmanned Aerial Vehicles have a lot of potentials in outdoor applications. However, uncertainties such as wind disturbances and mass change when performing some particular tasks, greatly affect their tracking performance. This paper presents a methodology using L1adaptive control to address some of the robustness issues of the quadcopter in outdoor flight which significantly improves the performance comparing to the baseline controller. Simulation and flight tests verify the potential of the presented controller.

This paper addresses the application of Model Predictive Control (MPC) approach for the Leader-followers formation flight problem. Under the robust decentralized unified MPC framework, the current collision avoidance scheme has been extended to take care of any shape and small pop-up obstacles. And in the heterogeneous Leader-follower situation, if the formation error term is added to the Leader's cost function, then the formation will be maintained. The simulation results show that the modified decentralized MPC framework can successfully achieve and keep collision-free formation flights.