Search Results - fengbo+ren

10 Results Sort By:
High Performance Computing Framework for Accelerating Sparse Cholesky Factorization on FPGAs
­Solving large symmetric sparse linear systems using sparse Cholesky factorization plays a pivotal role in many scientific computing and high-performance computing (HPC) applications. The existing computational solutions to sparse Cholesky factorization based on CPUs and GPUs suffer from very limited performance for to two primary reasons. First,...
Published: 2/23/2023   |   Inventor(s): Erfan Bank Tavakoli, Fengbo Ren, Michael Riera, Masudul Quraishi
Keywords(s): Algorithm Development, PS-Computing and Information Technology
Category(s): Computing & Information Technology, Physical Science
A Scalable Parameterized OpenCL-Defined Accelerator Architecture for Efficient Convolutional Neural Network (CNN) Inference on FPGAs
Background Convolutional neural networks (CNNs) are a class of deep neural network known for their superior extraction capability of shift/space invariant local features critical for high-level cognition tasks. CNNs are widely applied in image/video processing and computer vision tasks, including image/video recognition, object detection, and semantic...
Published: 2/23/2023   |   Inventor(s): Akshay Dua, Fengbo Ren
Keywords(s):  
Category(s): Physical Science, Computing & Information Technology, Imaging
Data-Driven Nonuniform Subsampling for Computation-Free Compressive Information Acquisition
Background The emergent Internet-of-Things (IoT) data explosion will inevitably exert a tremendous data transmission burden onto the wireless edge networks, which are often low-power wide-area networks (LPWAN) with very limited bandwidth. Thus, improving the bandwidth efficiency of wireless edge networks will be of great importance. In addition, the...
Published: 2/23/2023   |   Inventor(s): Zhikang Zhang, Fengbo Ren
Keywords(s):  
Category(s): Computing & Information Technology, Physical Science
Pruning Binary Neural Networks (BNNs) for Model Compression and Inference Acceleration
Background In the rapid evolution of deep learning, neural network models have been growing from several to over a hundred layers for handling more complex tasks. Network models are often trained with powerful GPUs in the cloud or on stand-alone servers, and the trained models are then deployed to certain hardware platforms for performing inference....
Published: 2/23/2023   |   Inventor(s): Yixing Li, Fengbo Ren
Keywords(s):  
Category(s): Physical Science, Computing & Information Technology
Generic Compression Ratio Adapter for End-to-End Data-Driven Image Compressive Sensing Reconstruction (EDCSR) Frameworks
Background Compressive sensing (CS) is a transformative sampling technique that allows sparse signals to be sampled in compressed form at a rate much lower than the Nyquist rate. In particular, end-to-end data-driven image compressive sensing reconstruction (EDCSR) frameworks achieve state-of-the-art reconstruction performance in terms of speed and...
Published: 2/23/2023   |   Inventor(s): Zhikang Zhang, Fengbo Ren, Kai Xu
Keywords(s):  
Category(s): Wireless & Networking, Physical Science, Computing & Information Technology
Framework for Extreme-Scale Heterogeneous High-Performance Computing (HPC)
In recent years, new accelerator choices for heterogeneous HPC have emerged, such as field-programmable gate arrays (FPGAs, considered as reconfigurable accelerators) and tensor processing units (TPUs, considered as domain-specific accelerators). Although these options offer flexible or customized hardware architectures with excellent capabilities for...
Published: 2/23/2023   |   Inventor(s): Michael Riera, Fengbo Ren, Masudul Quraishi, Erfan Bank Tavakoli
Keywords(s): Networks, IT, Software and Communication, Performance Optimization, PS-Computing and Information Technology, Software
Category(s): Physical Science, Computing & Information Technology
CSVideoNet: A Real-time End-to-end Learning Framework for High-frame-rate Video Compressive Sensing
CSVideoNet: A Real-time End-to-end Learning Framework for High-frame-rate Video Compressive Sensing Background High-frame-rate cameras are capable of capturing videos at frame rates over 100 frames per second (fps). These devices were originally developed to characterize rapid events. Some high-frame-rate cameras can record high resolution still images...
Published: 2/23/2023   |   Inventor(s): Fengbo Ren, Kai Xu
Keywords(s): Defense/military, Imaging
Category(s): Imaging, Applied Technologies
A Scalable Laplacian Pyramid Reconstructive Adversarial Network for Flexible Compressive Sensing Reconstruction
Background Compressive sensing (CS) is a transformative sampling technique that allows sparse signals to be sampled in compressed form at a rate much lower than the Nyquist rate. However, conventional CS reconstruction techniques, based on either convex optimization or iterative methods, come with three drawbacks in imaging applications: (1) High complexity...
Published: 2/23/2023   |   Inventor(s): Fengbo Ren, Kai Xu, Zhikang Zhang
Keywords(s):  
Category(s): Physical Science, Imaging, Computing & Information Technology
2013-960 Scalable Parameterized VLSI Architecture for Compressive Sensing Sparse Approximation
Summary Researchers in the UCLA Department of Electrical Engineering have developed a scalable and parameterized VLSI architecture for compressive sensing (CS) sparse approximation, allowing for energy-efficient, cost-effective, and real-time compressive-sampled data processing in wireless/mobile healthcare applications. Background Wireless healthcare...
Published: 7/19/2023   |   Inventor(s): Fengbo Ren, Dejan Markovic
Keywords(s):  
Category(s): Software & Algorithms > Communication & Networking
Body Voltage Sensing Based Short Pulse Reading Circuit For STT-RAM
SummaryUCLA researchers in the Department of Electrical & Computer Engineering have invented a novel circuit design that performs high speed and reliable data reading operations for resistive device-based memory applications.BackgroundReading circuits perform crucial data reading operations in memory applications, which has a fast-growing multi-billion...
Published: 7/19/2023   |   Inventor(s): Kang Wang, Chih Kong Yang, Dejan Markovic, Fengbo Ren
Keywords(s): Electronics & Semiconductors
Category(s): Electrical > Electronics & Semiconductors, Electrical > Electronics & Semiconductors > Circuits