623509 research outputs found
Sort by
Deep Learning-Based Noninvasive Screening of Type 2 Diabetes with Chest X-ray Images and Electronic Health Records
No full textThe imperative for early detection of type 2 diabetes mellitus (T2DM) is challenged by its asymptomatic onset and dependence on suboptimal clinical diagnostic tests, contributing to its widespread global prevalence. While research into noninvasive T2DM screening tools has advanced, conventional machine learning approaches remain limited to unimodal inputs due to extensive feature engineering requirements. In contrast, deep learning models can leverage multimodal data for a more holistic understanding of patients\u27 health conditions. However, the potential of chest X-ray (CXR) imaging, one of the most commonly performed medical procedures, remains underexplored. This study evaluates the integration of CXR images with other noninvasive data sources, including electronic health records (EHRs) and electrocardiography signals, for T2DM detection. Utilising datasets meticulously compiled from the MIMIC-IV databases, we investigated two deep fusion paradigms: an early fusion-based multimodal transformer and a modular joint fusion ResNet-LSTM architecture. The end-to-end trained ResNet-LSTM model achieved an AUROC of 0.86, surpassing the CXR-only baseline by 2.3% with just 9863 training samples. These findings demonstrate the diagnostic value of CXRs within multimodal frameworks for identifying at-risk individuals early. Additionally, the dataset preprocessing pipeline has also been released to support further research in this domain
Enhancing Off-Grid One-Bit DOA Estimation with Learning-Based Sparse Bayesian Approach for Non-Uniform Sparse Array
No full textThis paper tackles the challenge of one-bit off-grid direction of arrival (DOA) estimation in a single snapshot scenario based on a learning-based Bayesian approach. Firstly, we formulate the off-grid DOA estimation model, utilizing the first-order off-grid approximation, incorporating one-bit data quantization. Subsequently, we address this problem using the Sparse Bayesian based framework and solve iteratively. However, traditional Sparse Bayesian methods often face challenges such as high computational complexity and the need for extensive hyperparameter tuning. To balance estimation accuracy and computational efficiency, we propose a novel Learning-based Sparse Bayesian framework, which leverages an unrolled neural network architecture. This framework autonomously learns hyperparameters through supervised learning, offering more accurate off-grid DOA estimates and improved computational efficiency compared to some state-of-the-art methods. Furthermore, the proposed approach is applicable to both uniform linear arrays and non-uniform sparse arrays. Simulation results validate the effectiveness of the proposed framework.Proc. 58th Annual Asilomar Conference on Signals, Systems, and Computers (Asilomar), Pacific Grove, CA, Oct. 27 - Oct. 30, 202
On Scalable Design for User-Centric Multi-Modal Shared E-Mobility Systems using MILP and Modified Dijkstra\u27s Algorithm
No full textIn the rapidly evolving landscape of urban transportation, shared e-mobility services have emerged as a sustainable solution to meet growing demand for flexible, eco-friendly travel. However, the existing literature lacks a comprehensive multi-modal optimization framework with focus on user preferences and real-world constraints. This paper presents a multi-modal optimization framework for shared e-mobility, with a particular focus on e-mobility hubs (e-hubs) with micromobility. We propose and evaluate two approaches: a mixed-integer linear programming (MILP) solution, complemented by a heuristic graph reduction technique to manage computational complexity in scenarios with limited e-hubs, achieving a computational advantage of 93%, 72%, and 47% for 20, 50, and 100 e-hubs, respectively. Additionally, the modified Dijkstra\u27s algorithm offers a more scalable, real-time alternative for larger e-hub networks, with median execution times consistently around 53 ms, regardless of the number of e-hubs. Thorough experimental evaluation on real-world map and simulated traffic data of Dublin City Centre reveals that both methods seamlessly adapt to practical considerations and constraints such as multi-modality, user-preferences and state of charge for different e-mobility tools. While MILP offers greater flexibility for incorporating additional objectives and constraints, the modified Dijkstra\u27s algorithm is better suited for large-scale, real-time applications due to its computational efficiency.The paper has been accepted by the 2025 IEEE Symposium Series on Computational Intelligenc
A Contextualized BERT model for Knowledge Graph Completion
No full textKnowledge graphs (KGs) are valuable for representing structured, interconnected information across domains, enabling tasks like semantic search, recommendation systems and inference. A pertinent challenge with KGs, however, is that many entities (i.e., heads, tails) or relationships are unknown. Knowledge Graph Completion (KGC) addresses this by predicting these missing nodes or links, enhancing the graph\u27s informational depth and utility. Traditional methods like TransE and ComplEx predict tail entities but struggle with unseen entities. Textual-based models leverage additional semantics but come with high computational costs, semantic inconsistencies, and data imbalance issues. Recent LLM-based models show improvement but overlook contextual information and rely heavily on entity descriptions. In this study, we introduce a contextualized BERT model for KGC that overcomes these limitations by utilizing the contextual information from neighbouring entities and relationships to predict tail entities. Our model eliminates the need for entity descriptions and negative triplet sampling, reducing computational demands while improving performance. Our model outperforms state-of-the-art methods on standard datasets, improving Hit@1 by 5.3% and 4.88% on FB15k-237 and WN18RR respectively, setting a new benchmark in KGC.MuslML Workshop, 38th Conference on Neural Information Processing Systems (NeurIPS 2024
PolyModel for Hedge Funds\u27 Portfolio Construction Using Machine Learning
No full textThe domain of hedge fund investments is undergoing significant transformation, influenced by the rapid expansion of data availability and the advancement of analytical technologies. This study explores the enhancement of hedge fund investment performance through the integration of machine learning techniques, the application of PolyModel feature selection, and the analysis of fund size. We address three critical questions: (1) the effect of machine learning on trading performance, (2) the role of PolyModel feature selection in fund selection and performance, and (3) the comparative reliability of larger versus smaller funds.
Our findings offer compelling insights. We observe that while machine learning techniques enhance cumulative returns, they also increase annual volatility, indicating variability in performance. PolyModel feature selection proves to be a robust strategy, with approaches that utilize a comprehensive set of features for fund selection outperforming more selective methodologies. Notably, Long-Term Stability (LTS) effectively manages portfolio volatility while delivering favorable returns. Contrary to popular belief, our results suggest that larger funds do not consistently yield better investment outcomes, challenging the assumption of their inherent reliability.
This research highlights the transformative impact of data-driven approaches in the hedge fund investment arena and provides valuable implications for investors and asset managers. By leveraging machine learning and PolyModel feature selection, investors can enhance portfolio optimization and reassess the dependability of larger funds, leading to more informed investment strategies
Reconfigurable Intelligent Surface-Aided Secure Integrated Radar and Communication Systems
No full textDespite the enhanced spectral efficiency brought by the integrated radar and communication technique, it poses significant risks to communication security when confronted with malicious radar targets. To address this issue, a reconfigurable intelligent surface (RIS)-aided transmission scheme is proposed to improve secure communication in two systems, i.e., the radar and communication co-existing (RCCE) system, where a single transmitter is utilized for both radar sensing and communication, and the dual-functional radar and communication (DFRC) system. At the design stage, optimization problems are formulated to maximize the secrecy rate while satisfying the radar detection constraint via joint active beamforming at the base station and passive beamforming of RIS in both systems. Particularly, a zero-forcing-based block coordinate descent (BCD) algorithm is developed for the RCCE system. Besides, the Dinkelbach method combined with semidefinite relaxation is employed for the DFRC system, and to further reduce the computational complexity, a Riemannian conjugate gradient-based alternating optimization algorithm is proposed. Moreover, the RIS-aided robust secure communication in the DFRC system is investigated by considering the eavesdropper\u27s imperfect channel state information (CSI), where a bounded uncertainty model is adopted to capture the angle error and fading channel error of the eavesdropper, and a tractable bound for their joint uncertainty is derived. Simulation results confirm the effectiveness of the developed RIS-aided transmission scheme to improve the secrecy rate even with the eavesdropper\u27s imperfect CSI, and comparisons between both systems reveal that the RCCE system can provide a higher secrecy rate than the DFRC system.15 pages, 7 figures; To be published on IEEE Transactions on Wireless Communication
HC-LLM: Historical-Constrained Large Language Models for Radiology Report Generation
No full textRadiology report generation (RRG) models typically focus on individual exams, often overlooking the integration of historical visual or textual data, which is crucial for patient follow-ups. Traditional methods usually struggle with long sequence dependencies when incorporating historical information, but large language models (LLMs) excel at in-context learning, making them well-suited for analyzing longitudinal medical data. In light of this, we propose a novel Historical-Constrained Large Language Models (HC-LLM) framework for RRG, empowering LLMs with longitudinal report generation capabilities by constraining the consistency and differences between longitudinal images and their corresponding reports. Specifically, our approach extracts both time-shared and time-specific features from longitudinal chest X-rays and diagnostic reports to capture disease progression. Then, we ensure consistent representation by applying intra-modality similarity constraints and aligning various features across modalities with multimodal contrastive and structural constraints. These combined constraints effectively guide the LLMs in generating diagnostic reports that accurately reflect the progression of the disease, achieving state-of-the-art results on the Longitudinal-MIMIC dataset. Notably, our approach performs well even without historical data during testing and can be easily adapted to other multimodal large models, enhancing its versatility.Accepted by AAAI202
A Note on Valid Inequalities for PageRank Optimization with Edge Selection Constraints
No full textCsáji, Jungers, and Blondel prove that while a PageRank optimization problem with edge selection constraints is NP-hard, it can be solved optimally in polynomial time for the unconstrained case. This theoretical result is accompanied by several observations, which we leverage to develop valid inequalities in polynomial time for this class of NP-hard problems
Deep Spectral Clustering via Joint Spectral Embedding and Kmeans
No full textSpectral clustering is a popular clustering method. It first maps data into the spectral embedding space and then uses Kmeans to find clusters. However, the two decoupled steps prohibit joint optimization for the optimal solution. In addition, it needs to construct the similarity graph for samples, which suffers from the curse of dimensionality when the data are high-dimensional. To address these two challenges, we introduce \textbf{D}eep \textbf{S}pectral \textbf{C}lustering (\textbf{DSC}), which consists of two main modules: the spectral embedding module and the greedy Kmeans module. The former module learns to efficiently embed raw samples into the spectral embedding space using deep neural networks and power iteration. The latter module improves the cluster structures of Kmeans on the learned spectral embeddings by a greedy optimization strategy, which iteratively reveals the direction of the worst cluster structures and optimizes embeddings in this direction. To jointly optimize spectral embeddings and clustering, we seamlessly integrate the two modules and optimize them in an end-to-end manner. Experimental results on seven real-world datasets demonstrate that DSC achieves state-of-the-art clustering performance
Efficient hybrid-functional-based force and stress calculations for periodic systems with thousands of atoms
No full textWe present an efficient linear-scaling algorithm for evaluating the analytical force and stress contributions derived from the exact-exchange energy, a key component in hybrid functional calculations. The algorithm, working equally well for molecular and periodic systems, is formulated within the framework of numerical atomic orbital (NAO) basis sets and takes advantage of the localized resolution-of-identity (LRI) technique for treating the two-electron Coulomb repulsion integrals. The linear-scaling behavior is realized by fully exploiting the sparsity of the expansion coefficients resulting from the strict locality of the NAOs and the LRI ansatz. Our implementation is massively parallel, and enables efficient structural relaxation based on hybrid density functionals for bulk materials containing thousands of atoms. In this work, we will present a detailed description of our algorithm and benchmark the performance of our implementation using illustrating examples. By optimizing the structures of the pristine and doped halide perovskite material CsSnI with different functionals, we find that in the presence of lattice strain, hybrid functionals provide a more accurate description of the stereochemical expression of the lone pair