检索结果-内蒙古大学图书馆

NiemaGraphGen: A memory-efficient global-scale contact network simulation toolkit

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Moshiri, Niema Department of Computer Science & Engineering UC San Diego La Jolla 92093 United States

Epidemic simulations require the ability to sample contact networks from various random graph models. Existing methods can simulate city-scale or even country-scale contact networks, but they are unable to feasibly simulate global-scale contact networks due to high memory consumption. NiemaGraphGen (NGG) is a memory-efficient graph generation tool that enables the simulation of global-scale contact networks. NGG avoids storing the entire graph in memory and is instead intended to be used in a data streaming pipeline, resulting in memory consumption that is orders of magnitude smaller than existing tools. NGG provides a massively-scalable solution for simulating social contact networks, enabling global-scale epidemic simulation studies. © 2022, CC BY.

关键词： Epidemiology

SSCAE: A Neuromorphic SNN Autoencoder for sc-RNA-seq Dimensionality Reduction

学校读者我要写书评

暂无评论

SSCAE: A Neuromorphic SNN Autoencoder for sc-RNA-seq Dimensi...

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Tim Zhang Amirali Amirsoleimani Jason K. Eshraghian Mostafa Rahimi Azghadi Roman Genov Yu Xia Department of Bioengineering McGill University Montreal Canada Department of Electrical Engineering and Computer Science York University Toronto ON Canada Department of Electrical and Computer Engineering UC Santa Cruz CA United States College of Science and Engineering James Cook University Australia Department of Electrical and Computer Engineering University of Toronto Toronto Canada

Single-cell RNA sequencing is an emerging technique in the field of biology that departs radically from the previous assumption of gene-expression homogeneity within a tissue. The large quantity of data generated by this technology enables discoveries of cellular biology and disease mechanics that were previously not possible, and calls for accurate, scalable, and efficient processing pipelines. In this work, we propose SSCAE (spiking single-cell autoencoder), a novel SNN-based autoencoder for sc-RNA-seq dimensionality reduction. We apply this architecture to a variety of datasets, and the results show that it can match and surpass the performance of current state-of-the-art techniques. Moreover, the potential of this technique lies in its ability to be scaled up and to take advantage of neuromorphic hardware, circumventing the memory bottleneck that currently limits the size of sequencing datasets that can be processed.

关键词：

Forecasting Aortic Pressure Cross-Cohort with Deep Sequence Models

学校读者我要写书评

暂无评论

Forecasting Aortic Pressure Cross-Cohort with Deep Sequence ...

computers in Cardiology (CinC)

作者： Alan Li Zihao Zhou Elise Jortberg Rose Yu Department of Computer Science and Engineering UC San Diego USA Abiomed Danvers MA USA

Mean aortic pressure (MAP) is a primary measurement for monitoring blood and O 2 delivery to major organs. Prolonged periods of hypotension, low MAP, lead to low tissue perfusion and subsequent end organ damage. Patients on mechanical circulatory support (MCS) devices, such as the Impella CP, are managed to maintain sufficient MAP for end-organ perfusion. Forecasting MAP is important for early warning of clinically concerning events, including hypotension and instability as well as device weaning. Patients presenting with cardiogenic shock as a result of acute myocardial infarction (AMI/CGS) have increased hemodynamic instability when compared to patients undergoing high-risk percutaneous coronary interventions (HRPCI). Existing deep sequence models for fore-casting often focus on the same patient cohort and cannot generalize across cohorts. In this paper, we examine how deep sequence models respond to the distribution shift of the MAP across the MCS patient cohorts during forecasting. We propose conditional RNN, a deep sequence model that learns to adapt to a different cohort by conditioning on time-invariant cohort features. Our proposed model improves the forecasting performance, achieving a 5.2 mmHg – 6.1 mmHg RMSE for cross-cohort patients.

关键词： Adaptation models Electric shock Predictive models Myocardium Pressure measurement Hemodynamics Cardiology

Ultra-Thinned Individual SOI die ACF Fc Bonded on Rigid and Flex PCB 23

学校读者我要写书评

暂无评论

Ultra-Thinned Individual SOI die ACF Fc Bonded on Rigid and ...

23rd European Microelectronics and Packaging Conference and Exhibition, EMPC 2021

作者： Stoukatch, Serguei André, Nicolas Dupont, Francois Redouté, Jean-Michel Flandre, Denis Microsys lab Department of Electrical Engineering and Computer Science Liege University Seraing Belgium UC Louvain Louvain-la-Neuve Belgium

ISBN: (纸本)9780956808677

We have developed a straightforward die-level thinning process suitable for Silicon-On-Insulator (SOI) dies. The process has been demonstrated on SOI CMOS die assembled on rigid and flexible PCBs using previously-developed anisotropic conductive adhesive flip-chip method. Unlike standard wafer-level thinning processes, in the demonstrated process the full thickness SOI die is directly mounted on PCB and after that thinned. The demonstrated process is simple and robust;it comprises fewer process steps compared to conventional die thinning process. The ultra-thinning process has no effects on the assembly integrity and device performance. © 2021 2021 23rd European Microelectronics and Packaging Conference and Exhibition, EMPC 2021. All rights reserved.

关键词： Dies

AIoT and VR-Based Technology for Robots Control in Critical Safety Environments: Challenges and Opportunities 25

学校读者我要写书评

暂无评论

AIoT and VR-Based Technology for Robots Control in Critical ...

25th IEEE International Conference on Industrial Technology, ICIT 2024

作者： Ouyang, Yuanbing Luo, Yiming Ji, Ran Wu, Di Wang, Ding Wang, Xiaofen Wang, Hao Pan, Yushan Xidian University China Xi'an Jiaotong - Liverpool University Xi'an China School Of Cyber Engineering Xidian University Xi'an China School Of Mathematics Uc Irvine CA United States Ntnu Department Of Ict Aalesund Norway Google Ai Google Research GA United States School Of Computer Science Uestc Chengdu China Xjtlu Suzhou China

ISBN: (纸本)9798350340266

This study explores the integration of Artificial In-telligence of Things (AIoT) and Virtual Reality (VR) in research dedicated to operations within critical safety environments. The discussion commences by examining intentional interventions that may impact the interaction between remote operators and objects through the use of AIoT and VR devices as control interfaces. Following this, a detailed exposition of the tools used in AIoT-VR interfaces is presented, with a specific focus on applications in safety-critical scenarios. Subsequently, we delve into various heuristic design, implementation, and reporting methods employed in AIoT-VR-based research. This includes considerations such as participant selection, activity and task settings, and the careful selection and evaluation of experimental parameters. The primary aim of this paper is to provide guidance for enhancing the quality of AIoT-VR research in safety-critical operations. This is achieved through aligning with existing research backgrounds via comprehensive discussions and examples. The paper encourages researchers to adopt flexible approaches in applying context-specific designs and explorations in AIoT-VR studies for safety-critical operations. The anticipated impact of this review includes the practical advancement of AIoT-VR research within safety-critical studies, while simultaneously reinforcing the importance of user-centric considerations in engineering projects. © 2024 IEEE.

关键词： Virtual reality

Fast multiplication of random dense matrices with fixed sparse matrices

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Liang, Tianyu Murray, Riley Buluç, Aydın Demmel, James Electrical Engineering and Computer Science Department UC Berkeley United States Computational Research Division Lawrence Berkeley National Lab United States International Computer Science Institute United States Sandia National Laboratories

This work focuses on accelerating the multiplication of a dense random matrix with a (fixed) sparse matrix, which is frequently used in sketching algorithms. We develop a novel scheme that takes advantage of blocking and recomputation (on-the-fly random number generation) to accelerate this operation. The techniques we propose decrease memory movement, thereby increasing the algorithm's parallel scalability in shared memory architectures. On the Intel Frontera architecture, our algorithm can achieve 2x speedups over libraries such as Eigen and Intel MKL on some examples. In addition, with 32 threads, we can obtain a parallel efficiency of up to 45%. We also present a theoretical analysis for the memory movement lower bound of our algorithm, showing that under mild assumptions, it's possible to beat the data movement lower bound of general matrix-matrix multiply (GEMM) by a factor of √M, where M is the cache size. Finally, we incorporate our sketching method into a randomized algorithm for overdetermined least squares with sparse data matrices. Our results are competitive with SuiteSparse for highly overdetermined problems;in some cases, we obtain a speedup of 10x over SuiteSparse. Copyright © 2023, The Authors. All rights reserved.

关键词： Random number generation

Non-stationary Reinforcement Learning under General Function Approximation

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Feng, Songtao Yin, Ming Huang, Ruiquan Wang, Yu-Xiang Yang, Jing Liang, Yingbin Department of Electrical and Computer Engineering The Ohio State University OH43210 United States Department of Computer Science UC Santa Barbara CA93106 United States School of Electrical Engineering and Computer Science The Pennsylvania State University University ParkPA16802 United States

General function approximation is a powerful tool to handle large state and action spaces in a broad range of reinforcement learning (RL) scenarios. However, theoretical understanding of non-stationary MDPs with general function approximation is still limited. In this paper, we make the first such an attempt. We first propose a new complexity metric called dynamic Bellman Eluder (DBE) dimension for non-stationary MDPs, which subsumes majority of existing tractable RL problems in static MDPs as well as non-stationary MDPs. Based on the proposed complexity metric, we propose a novel confidence-set based model-free algorithm called SW-OPEA, which features a sliding window mechanism and a new confidence set design for non-stationary MDPs. We then establish an upper bound on the dynamic regret for the proposed algorithm, and show that SW-OPEA is provably efficient as long as the variation budget is not significantly large. We further demonstrate via examples of non-stationary linear and tabular MDPs that our algorithm performs better in small variation budget scenario than the existing ucB-type algorithms. To the best of our knowledge, this is the first dynamic regret analysis in nonstationary MDPs with general function approximation. Copyright © 2023, The Authors. All rights reserved.

关键词： Reinforcement learning

A Learning-based Quadcopter Controller with Extreme Adaptation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Dingqi Loquercio, Antonio Tang, Jerry Wang, Ting-Hao Malik, Jitendra Mueller, Mark W. High Performance Robotics Lab Dept. of Mechanical Engineering UC Berkeley United States The University of Pennsylvania United States Dept. of Electrical Engineering and Computer Science University of California at Berkeley United States

This paper introduces a learning-based low-level controller for quadcopters, which adaptively controls quadcopters with significant variations in mass, size, and actuator capabilities. Our approach leverages a combination of imitation learning and reinforcement learning, creating a fast-adapting and general control framework for quadcopters that eliminates the need for precise model estimation or manual tuning. The controller estimates a latent representation of the vehicle’s system parameters from sensor-action history, enabling it to adapt swiftly to diverse dynamics. Extensive evaluations in simulation demonstrate the controller’s ability to generalize to unseen quadcopter parameters, with an adaptation range up to 16 times broader than the training set. In real-world tests, the controller is successfully deployed on quadcopters with mass differences of 3.7 times and propeller constants varying by more than 100 times, while also showing rapid adaptation to disturbances such as off-center payloads and motor failures. These results highlight the potential of our controller to simplify the design process and enhance the reliability of autonomous drone operations in unpredictable environments. Video and code are at: https://***/muellerlab/xadapt_ctrl © 2024, CC BY.

关键词： Reinforcement learning

Mechanism of feature learning in convolutional neural networks

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Beaglehole, Daniel Radhakrishnan, Adityanarayanan Pandit, Parthe Belkin, Mikhail Halıcıoğlu Data Science Institute UC San Diego United States Computer Science and Engineering UC San Diego United States Massachusetts Institute of Technology United States Broad Institute of MIT and Harvard United States

Understanding the mechanism of how convolutional neural networks learn features from image data is a fundamental problem in machine learning and computer vision. In this work, we identify such a mechanism. We posit the Convolutional Neural Feature Ansatz, which states that covariances of filters in any convolutional layer are proportional to the average gradient outer product (AGOP) taken with respect to patches of the input to that layer. We present extensive empirical evidence for our ansatz, including identifying high correlation between covariances of filters and patch-based AGOPs for convolutional layers in standard neural architectures, such as AlexNet, VGG, and ResNets pre-trained on ImageNet. We also provide supporting theoretical evidence. We then demonstrate the generality of our result by using the patch-based AGOP to enable deep feature learning in convolutional kernel machines. We refer to the resulting algorithm as (Deep) ConvRFM and show that our algorithm recovers similar features to deep convolutional networks including the notable emergence of edge detectors. Moreover, we find that Deep ConvRFM overcomes previously identified limitations of convolutional kernels, such as their inability to adapt to local signals in images and, as a result, leads to sizable performance improvement over fixed convolutional kernels. © 2023, CC0.

关键词： Convolutional neural networks