检索结果-内蒙古大学图书馆

Graph Coordinates and Conventional Neural Networks - An Alternative for Graph Neural Networks

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Qin, Zheyi Paffenroth, Randy Jayasumana, Anura P. Colorado State University Fort Collins United States Department of Data Science Program Worcester Polytechnic Institute WorcesterMA United States

Graph-based data present unique challenges and opportunities for machine learning. Graph Neural Networks (GNNs), and especially those algorithms that capture graph topology through message passing for neighborhood aggregation, have been a leading solution. However, these networks often require substantial computational resources and may not optimally leverage the information contained in the graph’s topology, particularly for large-scale or complex graphs. We propose Topology Coordinate Neural Network (TCNN) and Directional Virtual Coordinate Neural Network (DVCNN) as novel and efficient alternatives to message passing GNNs, that directly leverage the graph’s topology, sidestepping the computational challenges presented by competing algorithms. Our proposed methods can be viewed as a reprise of classic techniques for graph embedding for neural network feature engineering, but they are novel in that our embedding techniques leverage ideas in Graph Coordinates (GC) that are lacking in current practice. Experimental results, benchmarked against the Open Graph Benchmark Leaderboard, demonstrate that TCNN and DVCNN achieve competitive or superior performance to message passing GNNs. For similar levels of accuracy and ROC-AUC, TCNN and DVCNN need far fewer trainable parameters than contenders of the OGBN Leaderboard. The proposed TCNN architecture requires fewer parameters than any neural network method currently listed in the OGBN Leaderboard for both OGBN-Proteins and OGBN-Products datasets. Conversely, our methods achieve higher performance for a similar number of trainable parameters. These results hold across diverse datasets and edge features, underscoring the robustness and generalizability of our methods. By providing an efficient and effective alternative to message passing GNNs, our work expands the toolbox of techniques for graph-based machine learning. A significantly lower number of tunable parameters for a given evaluation metric makes TCNN and DVCNN especia

关键词： Topology

STAR: Stepwise Task Augmentation and Relation Learning for Aspect Sentiment Quad Prediction

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Lai, Wenna Xie, Haoran Xu, Guandong Li, Qing Department of Computing Hong Kong Polytechnic University Hong Kong School of Data Science Lingnan University Hong Kong School of Computer Science Data Science Institute University of Technology Sydney SydneyNSW Australia Department of Computing The Hong Kong Polytechnic University Hong Kong

Aspect-based sentiment analysis (ABSA) aims to identify four sentiment elements, including aspect term, aspect category, opinion term, and sentiment polarity. These elements construct the complete picture of sentiments. The most challenging task, aspect sentiment quad prediction (ASQP), predicts these elements simultaneously, hindered by difficulties in accurately coupling different sentiment elements. A key challenge is insufficient annotated data that limits the capability of models in semantic understanding and reasoning about quad prediction. To address this, we propose stepwise task augmentation and relation learning (STAR), a strategy inspired by human reasoning. STAR constructs auxiliary data to learn quadruple relationships incrementally by augmenting with pairwise and overall relation tasks derived from training data. By encouraging the model to infer causal relationships among sentiment elements without requiring additional annotations, STAR effectively enhances quad prediction. Extensive experiments demonstrate the proposed STAR exhibits superior performance on four benchmark datasets. © 2025, CC BY-NC-SA.

关键词： Prediction models

RVISA: Reasoning and Verification for Implicit Sentiment Analysis

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Lai, Wenna Xie, Haoran Xu, Guandong Li, Qing Department of Computing Hong Kong Polytechnic University Hong Kong School of Computer Science The Data Science Institute University of Technology Sydney Australia School of Data Science Lingnan University Hong Kong School of Computer Science The Data Science Institute University of Technology Sydney SydneyNSW2007 Australia Education University of Hong Kong Hong Kong Department of Computing Hong Kong Polytechnic University Hong Kong

With an increasing social demand for fine-grained sentiment analysis (SA), implicit sentiment analysis (ISA) poses a significant challenge with the absence of salient cue words in expressions. It necessitates reliable reasoning to understand how the sentiment is aroused and thus determine implicit sentiments. In the era of Large Language Models (LLMs), Encoder-Decoder (ED) LLMs have gained popularity to serve as backbone models for SA applications, considering impressive text comprehension and reasoning ability among diverse tasks. On the other hand, Decoder-only (DO) LLMs exhibit superior natural language generation and in-context learning capabilities. However, their responses may contain misleading or inaccurate information. To identify implicit sentiment with reliable reasoning, this study proposes RVISA, a two-stage reasoning framework that harnesses the generation ability of DO LLMs and the reasoning ability of ED LLMs to train an enhanced reasoner. Specifically, we adopt three-hop reasoning prompting to explicitly furnish sentiment elements as cues. The generated rationales are utilized to fine-tune an ED LLM into a skilled reasoner. Additionally, we develop a straightforward yet effective verification mechanism to ensure the reliability of the reasoning learning. We evaluated the proposed method on two benchmark datasets and achieved state-of-the-art results in ISA performance. © 2024, CC BY-NC-SA.

关键词： Sentiment analysis

DOGMH: A Diffusion and Optimization Approach for Generative Molecular Health Informatics in Consumer Electronics

学校读者我要写书评

暂无评论

IEEE Transactions on Consumer Electronics 2025年

作者： Yang, Bolin Deng, Jie Liang, Xiaoxuan Chen, Zhenghan Wu, Ruoxue University of Chinese Academy of Sciences College of Life Sciences Beijing100081 China Nanyang Technological University College of Computing and Data Science Singapore117417 Singapore University of Massachusetts Amherst Department of ECE Amherst01002 United States Microsoft Bing Copilot Beijing100000 China Worcester Polytechnic Institute Department of Computer Science Worcester01609 United States

This is a task of computational chemistry and pharmaceutical development, with rapidly growing relevance for consumer healthcare electronics: deducing three-dimensional molecular structures from two-dimensional diagrams. In this paper, we introduce a new generative model, DOGMH, specifically tailored for applications on consumer electronic devices, which considerably increases the accuracy in this process. DOGMH borrows from the diffusion mechanisms of classical nonequilibrium thermodynamics and treats atoms as individual particles, excelling in the inversion of the diffusion process-turning noise into organized molecular structures. Our approach leverages a specially tailored bilevel optimization technique that ensures roto-translational invariance and allows for efficient processing on resource-constrained consumer devices. We show that DOGMH outperforms existing methods in terms of both conformational synthesis and distance distribution accuracy. DOGMH also provides the ability to create AI-enabled interactive 3D visualizations of molecular structures for consumer health education. Our work paves the way for advanced molecular informatics capabilities in consumer-grade devices and will potentially revolutionize personal healthcare management through AI-enhanced insights. By bridging the gap between complex molecular modeling and consumer electronics, DOGMH will open up a new frontier in intelligent, molecularly aware health devices accessible to the general public. © 2025 IEEE.

关键词： Diffusion

Adaptive Knowledge Graphs Enhance Medical Question Answering: Bridging the Gap Between LLMs and Evolving Medical Knowledge

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Rezaei, Mohammad Reza Fard, Reza Saadati Parker, Jayson L. Krishnan, Rahul G. Lankarany, Milad Department of Biomedical Engineering University of Toronto Canada Department of Computer Science Worcester Polytechnic Institute United States Department of Biology University of Toronto Mississauga Canada Department of Computer Science University of Toronto Canada Vector Institute Canada

Large Language Models (LLMs) have greatly advanced medical Question Answering (QA) by leveraging vast clinical data and medical literature. However, the rapid evolution of medical knowledge and the labor-intensive process of manually updating domain-specific resources can undermine the reliability of these systems. We address this challenge with Adaptive Medical Graph-RAG (AMG-RAG), a comprehensive framework that automates the construction and continuous updating of Medical Knowledge Graphs (MKGs), integrates Chain-of-Thought (CoT) reasoning, and retrieves current external evidence (e.g., PubMed, WikiSearch). By dynamically linking new findings and complex medical concepts, AMG-RAG not only boosts accuracy but also enhances interpretability for medical queries. Evaluations on the MEDQA and MEDMCQA benchmarks demonstrate the effectiveness of AMG-RAG, achieving an F1 score of 74.1% on MEDQA and an accuracy of 66.34% on MEDMCQA—surpassing both comparable models and those 10 to 100 times larger. Importantly, these improvements are achieved without increasing computational overhead, underscoring the critical impact of automated knowledge graph generation and external evidence retrieval in delivering up-to-date, trustworthy medical insights. © 2025, CC BY.

关键词： Knowledge graph

Vision-based FDM Printing for Fabricating Airtight Soft Actuators

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wu, Yijia Dai, Zilin Liu, Haotian Wang, Lehong Nemitz, Markus P. Department of Robotics Engineering Worcester Polytechnic Institute 100 Institute Road WorcesterMA01609 United States Department of Computer Science Worcester Polytechnic Institute 100 Institute Road WorcesterMA01609 United States Department of Mechanical and Materials Engineering Worcester Polytechnic Institute 100 Institute Road WorcesterMA01609 United States Department of Electrical and Computer Engineering Worcester Polytechnic Institute 100 Institute Road WorcesterMA01609 United States

Pneumatic soft robots are typically fabricated by molding, a manual fabrication process that requires skilled labor. Additive manufacturing has the potential to break this limitation and speed up the fabrication process but struggles with consistently producing high-quality prints. We propose a low-cost approach to improve the print quality of desktop fused deposition modeling by adding a webcam to the printer to monitor the printing process and detect and correct defects such as holes or gaps. We demonstrate that our approach improves the air-tightness of printed pneumatic actuators while reducing the need for fine-tuning printing parameters. Our approach presents a new option for robustly fabricating airtight, soft robotic actuators. Copyright © 2023, The Authors. All rights reserved.

关键词： Pneumatic actuators

Mobile Augmented Reality Framework with Fusional Localization and Pose Estimation

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Hou, Songlin Lin, Fangzhou Huang, Yunmei Peng, Zhe Xiao, Bin Department of Computer Science Worcester Polytechnic Institute Worcester01609 United States Department of Robotics Engineering Worcester Polytechnic Institute Worcester01609 United States Unprecedented-scale Data Analytics Center Tohoku University Sendai980-8577 Japan Department of Forestry and Natural Resources Purdue University West Lafayette47907 United States Department of Computing The Hong Kong Polytechnic University 999-077 Hong Kong

As a novel way of presenting information, augmented reality (AR) enables people to interact with the physical world in a direct and intuitive way. While there are some mobile AR products implemented with specific hardware at a high cost, the software approaches of AR implementation on mobile platforms(such as smartphones, tablet PC, etc.) are still far from practical use. GPS-based mobile AR systems usually perform poorly due to the inaccurate positioning in the indoor environment. Previous vision-based pose estimation methods need to continuously track predefined markers within a short distance, which greatly degrade user experience. This paper first conducts a comprehensive study of the state-of-the-art AR and localization systems on mobile platforms. Then, we propose an effective indoor mobile AR framework. In the framework, a fusional localization method and a new pose estimation implementation are developed to increase the overall matching rate and thus improving AR display accuracy. Experiments show that our framework has higher performance than approaches purely based on images or Wi-Fi signals. We achieve low average error distances (0.61-0.81m) and accurate matching rates (77%-82%) when the average sampling grid length is set to 0.5m. © 2025, CC0.

关键词： Wi-Fi

STREAM: Software Tool for Routing Efficiently Advanced Macrofluidics

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Wang, Lehong Kendre, Savita V. Liu, Haotian Nemitz, Markus P. Robotic Materials Group Department of Robotics Engineering Worcester Polytechnic Institute MA01609 United States Department of Computer Science Worcester Polytechnic Institute MA01609 United States Department of Electrical and Computer Engineering Worcester Polytechnic Institute MA01609 United States

The current fabrication and assembly of fluidic circuits for soft robots relies heavily on manual processes;as the complexity of fluidic circuits increases, manual assembly becomes increasingly arduous, error-prone, and time-consuming. We introduce a software tool that generates printable fluidic networks automatically. We provide a library of fluidic logic elements that are easily 3D printed from thermoplastic polyurethanes using Fused Deposition Modeling only. Our software tool and component library allow the development of arbitrary soft digital circuits. We demonstrate a variable frequency ring oscillator and a full adder. The simplicity of our approach using FDM printers only, democratizes fluidic circuit implementation beyond specialized laboratories. Our software is available on GitHub (https://***/roboticmaterialsgroup/FluidLogic) Copyright © 2023, The Authors. All rights reserved.

关键词： computer software

SF-DQN: Provable Knowledge Transfer using Successor Feature for Deep Reinforcement Learning 41

学校读者我要写书评

暂无评论

SF-DQN: Provable Knowledge Transfer using Successor Feature ...

41st International Conference on Machine Learning, ICML 2024

作者： Zhang, Shuai Fernando, Heshan Devaka Liu, Miao Murugesan, Keerthiram Lu, Songtao Chen, Pin-Yu Chen, Tianyi Wang, Meng Department of Data Science New Jersey Institute of Technology NewarkNJ United States Department of Electrical Computer and Systems Engineering Rensselaer Polytechnic Institute TroyNY United States IBM Thomas J.Watson Research Center IBM Research Yorktown HeightsNY United States

This paper studies the transfer reinforcement learning (RL) problem where multiple RL problems have different reward functions but share the same underlying transition *** this setting, the Q-function of each RL problem (task) can be decomposed into a successor feature (SF) and a reward mapping: the former characterizes the transition dynamics, and the latter characterizes the task-specific reward *** Q-function decomposition, coupled with a policy improvement operator known as generalized policy improvement (GPI), reduces the sample complexity of finding the optimal Q-function, and thus the SF & GPI framework exhibits promising empirical performance compared to traditional RL methods like ***, its theoretical foundations remain largely unestablished, especially when learning the successor features using deep neural networks (SF-DQN).This paper studies the provable knowledge transfer using SFs-DQN in transfer RL *** establish the first convergence analysis with provable generalization guarantees for SF-DQN with *** theory reveals that SF-DQN with GPI outperforms conventional RL approaches, such as deep Q-network, in terms of both faster convergence rate and better *** experiments on real and synthetic RL tasks support the superior performance of SF-DQN & GPI, aligning with our theoretical findings. Copyright 2024 by the author(s)

关键词： Deep reinforcement learning