检索结果-内蒙古大学图书馆

Opportunities and Challenges from Using Animal Videos in Reinforcement Learning for Navigation

IFAC-PapersOnLine 2023年第2期56卷 9056-9061页

作者： Vittorio Giammarino James Queeney Lucas C. Carstensen Michael E. Hasselmo Ioannis Ch Paschalidis Division of Systems Engineering Boston University Boston MA 02215 USA Center for Systems Neuroscience Department of Psychological and Brain Sciences and Graduate Program for Neuroscience Boston University Boston MA 02215 USA College of Engineering and Hariri Institute for Computing & Computational Science & Engineering Boston University Boston MA 02215 USA

We investigate the use of animal videos (observations) to improve Reinforcement Learning (RL) efficiency and performance in navigation tasks with sparse rewards. Motivated by theoretical considerations, we make use of weighted policy optimization for off-policy RL and describe the main challenges when learning from animal videos. We propose solutions and test our ideas on a 2D navigation task. We show how the use of animal videos improves performance over RL algorithms that do not leverage such observations.

关键词： Learning for Control Reinforcement Learning Deep Learning RL with demonstrations Imitation Learning Bio-inspired Control

来源：评论

学校读者我要写书评

暂无评论

An Efficient Novel Approach for Iris Recognition and Segmentation Based on the Utilization of Deep Learning

An Efficient Novel Approach for Iris Recognition and Segment...

引用

2023 International Conference on Evolutionary Algorithms and Soft computing Techniques, EASCT 2023

作者： Pasha, Syed Nawaz Anitha, D. Anbunathan, R. Gobinath, C. Madhura, G.K. Sr University School of Computer Science & Artificial Intelligence Warangal India School of Computing Srm Institute of Science and Technology Department of Computational Intelligence Kattankulathur India Sa Engineering College Department of Mathematics Chennai India Reva University School of Computing and Information Technology Bangalore India Nitte Meenakshi Institute of Technology Department of Ai and Ml Bengaluru India

ISBN: (纸本)9798350313413

Iris recognition stands out as one of the most stable and accurate biometric identification systems such as facial, voice, fingerprint, handwriting, and iris recognition itself. Deep Learning (DL) plays a pivotal role in enhancing security by leveraging Convolutional Neural Networks (CNNs) for iris recognition. The acquisition of iris images and real-time processing have become focal points demanding heightened attention in the realm of DL. Iris recognition is prized for its high level of protection, making it challenging to reconstruct iris data easily from the database. The CASIA-Iris Dataset serves as a benchmark to verify the uniqueness of iris patterns and evaluate the performance of iris recognition. However, due to limitations in iris image access, the application of CNNs in the segmentation approach remains critical for iris localization and pupil boundary identification, ensuring both efficiency and specificity in iris images. The results obtained from this study reveal that the proposed CNN, utilizing DL models, achieves an impressive accuracy of 99.89% on the CASIA-Iris dataset, better performance of other existing models, such as U-Net and VGG-Net, and ensuring accurate classification. © 2023 IEEE.

关键词： Contrast-Limited Adaptive Histogram Equalization Convolution Neural Network Deep Learning Iris Recognition and Support Vector Machine

来源：评论

学校读者我要写书评

暂无评论

Microbe-bridged disease-metabolite associations identification by heterogeneous graph fusion

引用

BRIEFINGS IN BIOINFORMATICS 2022年第6期23卷 bbac423-bbac423页

作者： Feng, Jitong Wu, Shengbo Yang, Hongpeng Ai, Chengwei Qiao, Jianjun Xu, Junhai Guo, Fei College of Intelligence and Computing Tianjin University Tianjin China School of Chemical Engineering and Technology Tianjin University Tianjin China Zhejiang Shaoxing Research Institute of Tianjin University Shaoxing China School of Computational Science and Engineering University of South Carolina Columbia U.S School of Computer Science and Engineering Central South University Changsha China

Motivation: Metabolomics has developed rapidly in recent years, and metabolism-related databases are also gradually constructed. Nowadays, more and more studies are being carried out on diverse microbes, metabolites and diseases. However, the logics of various associations among microbes, metabolites and diseases are limited understanding in the biomedicine of gut microbial system. The collection and analysis of relevant microbial bioinformation play an important role in the revelation of microbe-metabolite-disease associations. Therefore, the dataset that integrates multiple relationships and the method based on complex heterogeneous graphs need to be developed. Results: In this study, we integrated some databases and extracted a variety of associations data among microbes, metabolites and diseases. After obtaining the three interconnected bilateral association data (microbe-metabolite, metabolite-disease and disease-microbe), we considered building a heterogeneous graph to describe the association data. In our model, microbes were used as a bridge between diseases and metabolites. In order to fuse the information of disease-microbe-metabolite graph, we used the bipartite graph attention network on the disease-microbe and metabolite-microbe bipartite graph. The experimental results show that our model has good performance in the prediction of various disease-metabolite associations. Through the case study of type 2 diabetes mellitus, Parkinson's disease, inflammatory bowel disease and liver cirrhosis, it is noted that our proposed methodology are valuable for the mining of other associations and the prediction of biomarkers for different human diseases. Availability and implementation: https://***/Selenefreeze/***

来源：评论

学校读者我要写书评

暂无评论

Energy-Efficient Routing in Underwater Wireless Sensor Networks Through Cluster-Dragonfly Optimization 3

Energy-Efficient Routing in Underwater Wireless Sensor Netwo...

引用

3rd IEEE International Conference on Distributed computing and Electrical Circuits and Electronics, ICDCECE 2024

作者： Anitha, D. Husain, Saif O. Velpula, Srikanth Rajani, N. Dineshkumar, R. School Of Computing Srm Institute Of Science And Technology Department Of Computational Intelligence Tamilnadu Kattankulathur India College Of Technical Engineering The Islamic University Department Of Computers Techniques Engineering Najaf Iraq Sr University Department Of Electrical And Electronics Engineering Warangal India Nitte Meenakshi Institute Of Technology Dept.of Ece Bengaluru India Saveetha School Of Engineering Saveetha Institute Of Medical And Technical Sciences Saveetha University Department Of Electronics And Communication Engineering Chennai India

ISBN: (纸本)9798350318609

In this paper, develop a power control scheme for underwater wireless sensor networks that takes into account the impact of communication ranges on network performance, including energy utilization and delay overhead. The primary objective of our scheme is to assign a differentiated optimal communication radius for each sensor node, enhancing overall network efficiency. The proposed method utilizes energy-aware Cluster-Dragonfly Optimization (CDFO), but it might face challenges in handling large-scale optimization, and the scalability of the algorithm may be limited when dealing with high-dimensional or complex optimization. Although the proposed CDFO method exhibits high performance compared to existing methods, such as Self-Adaptive Glow Worm Swarm Optimization (SA-GWO), Improved Coyote Optimization Algorithm (ICOA), Hybrid Cat Cheetah Optimization algorithm (HCCOA), and Chaotic Search-and-Rescue Optimization (CSRO), it may still face challenges in terms of handling large-scale optimization. Nevertheless, the proposed CDFO yields impressive results, achieving a Packet Delivery Ratio (PDR) of 99%, Packet Loss Ratio (PLR) of 4%, Energy Consumption of 9%, and Alive Node count of 330. These outcomes outperform existing methods in the context of Underwater WSN. © 2024 IEEE.

关键词： Optimization

来源：评论

学校读者我要写书评

暂无评论

Finite Temperature Transition in Hyper Stealth Dark Matter using Möbius Domain Wall Fermions

arXiv

引用

arXiv 2025年

作者： Ayyar, Venkitesh Matsumoto, Nobuyuki Meyer, Aaron S. Park, Sungwoo Hariri Institute for Computing and Computational Science and Engineering Boston University BostonMA02215 United States Physical and Life Sciences Division Lawrence Livermore National Laboratory LivermoreCA94550 United States Nuclear Science Division Lawrence Berkeley National Laboratory BerkeleyCA94720 United States

The first-order confinement transition of a strongly coupled composite dark matter theory can provide a possible source of gravitational waves in the early universe. In this work, on behalf of the Lattice Strong Dynamics (LSD) Collaboration, we present our recent investigation on the finite temperature confinement transition of the one-flavor SU(4) dark gauge theory named Hyper Stealth Dark Matter (HSDM). The dark matter candidate in this theory is a composite bosonic baryon and can have a remarkably low mass of a few GeV. We expect the finite temperature transition to be first-order over at least in some finite range of fermionic masses and to be a potential source of observable gravitational radiation. The finite temperature simulation of one-flavor SU(4) is done by using Möbius Domain wall fermions. The order of the transition and its fermionic mass dependence are explored by monitoring the Polyakov loop, chiral condensate and topological charge using three lattice volumes at Nt = 8. © 2025, CC BY.

关键词： Fermions

来源：评论

学校读者我要写书评

暂无评论

Energy Efficient Mathematical Model for Cyber-Physical Systems: A Model for Industrial Internet of Things

引用

SN Computer science 2023年第5期4卷 1-7页

作者： Veerachamy, Ramachandran Rao, Ganga Rama Koteswara Arivanantham, Vishnu Priya Kuppusamy, Sangeetha Velayutham, Priya Govindaraj, Rajeshkumar Department of Computer Science and Engineering GITAM School of Computing GITAM Deemed to be University Bengaluru 561203 India Department of Computer Science and Information Technology Koneru Lakshmaiah Education Foundation Andhra Pradesh Vaddeswaram 520002 India Department of Computational Intelligence School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Vellore 632014 India Department of Computer Science and Engineering Kongu Engineering College Tamil Nadu Perundurai 638060 India Department of Computer Science and Engineering Paavai Engineering College Tamil Nadu Namakkal 637018 India Department of Information Science and Engineering Bannari Amman Institute of Technology Tamil Nadu Sathyamagalam 638401 India

Proliferated usage of computer devices, smart objects, and electronics significantly impacts the atmosphere, which is inevitable. The existing context has engineered schemes, or Cyber-Physical Systems (CPS), that integrate physical and computation components. The extensive usage of devices has utilized more energy, creating complexity in environmental handling. The problem is improving an automotive function’s reliability, application, or functionality with time and energy constraints. By applying smart technologies, namely Deep Learning (DL) and Machine Learning (ML) in CPS and the Internet of Things (IoT), enduring the utilization of these intensive DL-based approaches has a negative effect on resource utilization. The proposed Energy Efficient Mathematical Model (EEMM) is designed with a mathematical model for IIoT, which guides in solving the drawback of the existing system. The mathematical model integrated CPS-based proposed EEMM has outperformed the existing state-of-art technique. The proposed EEMM attains minimal Energy Consumption (EC) of 521 Jules and the highest reliability rate of 0.987. The proposed EEMM in this research has attained minimal energy usage of 0.987 i for 100 tasks, which is minimal than other existing techniques. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Cyber-attacks Energy IIoT IoT Mathematical model Network Smart objects

来源：评论

学校读者我要写书评

暂无评论

Projected gradient descent algorithm for ab initio crystal structure relaxation under a fixed unit cell volume

引用

Physical Review B 2024年第22期109卷 224108-224108页

作者： Yukuan Hu Junlei Yin Xingyu Gao Xin Liu Haifeng Song Institute of Computational Mathematics and Scientific/Engineering Computing Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing 100190 China School of Mathematical Sciences Innovation Center NPU Chongqing Chongqing 401135 China National Key Laboratory of Computational Physics

This paper is concerned with ab initio crystal structure relaxation under a fixed unit cell volume, which is a step in calculating the static equations of state and forms the basis of thermodynamic property calculations for materials. The task can be formulated as an energy minimization with a determinant constraint. Widely used line minimization-based methods (e.g., conjugate gradient method) lack both efficiency and convergence guarantees due to the nonconvex nature of the determinant constraint as well as the significant differences in the curvatures of the potential energy surface with respect to atomic and lattice components. To this end, we propose a projected gradient descent algorithm named PANBB. It is equipped with (i) search direction projections for lattice vectors, (ii) distinct curvature-aware initial trial step sizes for atomic and lattice updates, and (iii) a nonrestrictive line minimization criterion as the stopping rule for the inner loop. It can be proved that PANBB favors theoretical convergence to equilibrium states. Across a benchmark set containing 223 structures from various categories, PANBB achieves average speedup factors of approximately 1.41 and 1.45 over the conjugate gradient method and direct inversion in the iterative subspace implemented in off-the-shelf simulation software, respectively. Moreover, it normally converges on all the systems, manifesting its robustness. As an application, we calculate the static equations of state for the high-entropy alloy AlCoCrFeNi, which remains elusive owing to 160 atoms representing both chemical and magnetic disorder and the strong local lattice distortion. The results are consistent with the previous calculations and are further validated by experimental thermodynamic data.

关键词： Crystal structure Structural properties Density functional theory

来源：评论

学校读者我要写书评

暂无评论

SPARSE RECOVERY OF ELLIPTIC SOLVERS FROM MATRIX-VECTOR PRODUCTS

引用

SIAM Journal on Scientific computing 2024年第2期46卷 A998-A1025页

作者： Schäfer, Florian Owhadi, Houman Computational Science and Engineering Georgia Institute of Technology AtlantaGA30332 United States Department of Computing + Mathematical Sciences California Institute of Technology PasadenaCA91125 United States

In this work, we show that solvers of elliptic boundary value problems in d dimensions can be approximated to accuracy ϵ from only (log(N) logd(N/ϵ)) matrix-vector products with carefully chosen vectors (right-hand sides). The solver is only accessed as a black box, and the underlying operator may be unknown and of an arbitrarily high order. Our algorithm (1) has complexity (N log2(N) log2d(N/ϵ)) and represents the solution operator as a sparse Cholesky factorization with (N log(N) logd(N/ϵ)) nonzero entries, (2) allows for embarrassingly parallel evaluation of the solution operator and the computation of its log-determinant, (3) allows for (log(N) logd(N/ϵ)) complexity computation of individual entries of the matrix representation of the solver that, in turn, enables its recompression to an (N logd(N/ϵ)) complexity representation. As a byproduct, our compression scheme produces a homogenized solution operator with near-optimal approximation accuracy. By polynomial approximation, we can also approximate the continuous Green's function (in operator and Hilbert-Schmidt norm) to accuracy ϵ from (log1+d(ϵ 1)) solutions of the PDE. We include rigorous proofs of these results. To the best of our knowledge, our algorithm achieves the best known trade-off between accuracy ϵ and the number of required matrix-vector products. © 2024 Society for Industrial and Applied Mathematics Publications. All rights reserved.

关键词： Principal component analysis

来源：评论

学校读者我要写书评

暂无评论

Optimized Predictive Storage Resource Allocation Through Neural Forecasting for Cloud Environments

Optimized Predictive Storage Resource Allocation Through Neu...

引用

2024 International Conference on Innovative computing, Intelligent Communication and Smart Electrical Systems, ICSES 2024

作者： Sridevi Ponmalar, P. Adithya Pothan Raj, V. Jebaseelan, S. D. Sundarsingh Phade, Gayatri Srithar, V. Harikrishnan, R. School of computing Srm Institute of Science and Technology Department of Computational Intelligence Kattankulathur Campus Chennai India Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Dept. of Cse Chennai India Sathyabama Institute of Science and Technology Dept of Eee Chennai119 India Sandip Institute of Technology & Research Centre Nashik India PeriyarManiammai Institute of Science & Technology Department of Computer Applications Thanjavur India Symbiosis Institute of Technology Symbiosis International Deemed University Pune Campus Pune412115 India

ISBN: (纸本)9798331543617

The demand for efficient and scalable management of storage resources is growing enormously with the rapid growth in data-intensive applications within the cloud environment. Optimized PSRANF is the state-of-the-art approach towards dynamic storage resource allocation, based on the forecast of future demands in storage. PSRANF utilizes state-of-the-art neural forecasting with Long Short-Term Memory (LSTM) networks and Transformer designs to analyze the pattern of historical storage usage for highly accurate future need predictions. By proactively adjusting storage capacity to these predictions, PSRANF strikes an optimal balance between performance and cost by reserving just-in-time resources. It ensures that there is no risk of over-provisioning, which would be too costly. In the same breath, the risk of under-provisioning, which leads to performance bottlenecks, is avoided. What is more, the system will apply prioritization, placing the data of higher priority or frequently accessed data into faster storage tiers of a higher class of performance and less important classes of data into cost-effective storage. The adaptive feedback loop continuously refines the forecasting model, which, over time, improves accuracy and allows the system to handle changes in workload patterns. PSRANF combines neural forecasting with resource allocation in such a way that is cost-effective;thus, it is very well placed for cloud environments where the demand for storage is dynamic in nature and requires intelligent data-driven approaches. The AWS implementation of PSRANF has turned out to be quite a helpful tool in optimizing the storage resource allocation process. While it is able to predict 98.5% of the storage needs correctly by achieving a cost saving of up to 18.7%, managing data from 97.8% to 99.2%, and resources from 95.4%, it's robust for improvements in cloud storage management and operational efficiency. © 2024 IEEE.

关键词： Storage efficiency

来源：评论

学校读者我要写书评

暂无评论

A DYNAMICAL VARIABLE-SEPARATION METHOD FOR PARAMETER-DEPENDENT DYNAMICAL SYSTEMS

引用

SIAM Journal on Scientific computing 2025年第3期47卷 A1783-A1808页

作者： Chen, Liang Chen, Yaru Li, Qiuqi Zhou, Tao School of Mathematics Hunan University Changsha410082 China LSEC Institute of Computational Mathematics and Scientific/Engineering Computing Academy of Mathematics and Systems Science Chinese Academy of Sciences Beijing100190 China

This paper proposes a dynamical variable-separation method for solving parameter-dependent dynamical systems. To achieve this, we establish a dynamical low-rank approximation for the solutions of these dynamical systems by successively enriching each term in the reduced basis functions via a greedy algorithm. This enables us to reformulate the problem to two decoupled evolution equations at each enrichment step, one of which is a parameter-independent PDE and the other a parameter-dependent ODE. These equations are directly derived from the original dynamical system and the previously separate representation terms. Moreover, the computational process of the proposed method can be split into an offline stage, in which the reduced basis functions are constructed, and an online stage, in which the efficient low-rank representation of the solution is employed. The proposed dynamical variable-separation method is capable of offering reduced computational complexity and enhanced efficiency compared to many existing low-rank separation techniques. Finally, we present various numerical results for linear/nonlinear parameter-dependent dynamical systems to demonstrate the effectiveness of the proposed method. © 2025 Society for Industrial and Applied Mathematics.

关键词： Numerical methods

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：