Accelerating LSTM-based High-Rate Dynamic System Models

E Kabir, D Coble, JN Satme… - … Conference on Field …, 2023 - ieeexplore.ieee.org
In this paper, we evaluate the use of a trained Long Short-Term Memory (LSTM) network as
a surrogate for a Euler-Bernoulli beam model, and then we describe and characterize an …

FPGA Processor In Memory Architectures (PIMs): Overlay or Overhaul?

MDA Kabir, E Kabir, J Hollis… - … Conference on Field …, 2023 - ieeexplore.ieee.org
The dominance of machine learning and the ending of Moore's law have renewed interests
in Processor in Memory (PIM) architectures. This interest has produced several recent …

ReMoDeL-FPGA: Reconfigurable Memory-centric Array Processor Architecture for Deep-Learning Acceleration on FPGA

MA Kabir - 2024 - search.proquest.com
Deep-Learning has become a dominant computing paradigm across a broad range of
application domains. Different architectures of Deep-Networks like CNN, MLP, and RNN …

A Memory-Centric Customizable Domain-Specific FPGA Overlay for Accelerating Machine Learning Applications

A Panahi - 2022 - search.proquest.com
Low latency inferencing is of paramount importance to a wide range of real time and user
facing Machine Learning (ML) applications. Field Programmable Gate Arrays (FPGAs) offer …

Online Backpropagation of Recurrent Neural Network for Forecasting Nonstationary Structural Responses

Z Thompson, A Vereen, A Downey, JD Bakos… - Society for Experimental …, 2023 - Springer
Online time series forecasting of dynamic structural responses subjected to nonstationary
inputs enables the development of prognostic models used to compute the remaining useful …