paper-with-me

Papers

Recurrent Neural Networks Hardware Implementation on FPGA

2015-11-17 · Andre Xian Ming Chang, Berin Martini, Eugenio Culurciello

Recurrent Neural Networks (RNNs) have the ability to retain memory and learn data sequences. Due to the recurrent nature of RNNs, it is sometimes hard to parallelize all its computations on conventional hardware. CPUs do not currently offer large parallelism, while GPUs offer limited parallelism due to sequential components of RNN models. In this paper we present a hardware implementation of Long-Short Term Memory (LSTM) recurrent network on the programmable logic Zynq 7020 FPGA from Xilinx. We implemented a RNN with $2$ layers and $128$ hidden units in hardware and it has been tested using a character level language model. The implementation is more than $21\times$ faster than the ARM CPU embedded on the Zynq 7020 FPGA. This work can potentially evolve to a RNN co-processor for future mobile devices.

📄 PDF Abstract BibTeX arXiv:1511.05552

Code (1)

karpathy/char-rnn 공식 구현 torch

Tasks

CPULanguage ModelingLanguage Modelling

Similar Papers 제목 키워드 기반

Optimizing Bayesian Recurrent Neural Networks on an FPGA-based Accelerator

2021-06-04 · Martin Ferianc, Zhiqiang Que, Hongxiang Fan, Wayne Luk 외

Neural networks have demonstrated their outstanding performance in a wide range of tasks. Specifically recurrent architectures based on long-short term memory (LSTM) cells have manifested excellent capability to model ti…

GPUTime Series Analysis

E-RNN: Design Optimization for Efficient Recurrent Neural Networks in FPGAs

2018-12-12 · Zhe Li, Caiwen Ding, Siyue Wang, Wujie Wen 외

Recurrent Neural Networks (RNNs) are becoming increasingly important for time series-related applications which require efficient and real-time implementations. The two major types are Long Short-Term Memory (LSTM) and G…

Automatic Speech RecognitionAutomatic Speech Recognition (ASR)Quantizationspeech-recognition+2

Hardware-Efficient FPGA Implementation of Sigmoid Function Using Mixed-Radix Hyperbolic Rotation CORDIC

2026-04-26 · Chintan Panchal, Ankur Changela, Mohendra Roy arxiv

Efficient hardware implementation of nonlinear activation functions is a crucial task in deploying artificial neural networks on resource-constrained and edge devices such as Field-Programmable Gate Arrays (FPGAs). The s…

Binary Classification

FINN-L: Library Extensions and Design Trade-off Analysis for Variable Precision LSTM Networks on FPGAs

2018-07-11 · Vladimir Rybalkin, Alessandro Pappalardo, Muhammad Mohsin Ghaffar, Giulio Gambardella 외

It is well known that many types of artificial neural networks, including recurrent networks, can achieve a high classification accuracy even with low-precision weights and activations. The reduction in precision general…

Optical Character RecognitionOptical Character Recognition (OCR)Quantization

Hardware-accelerated graph neural networks: an alternative approach for neuromorphic event-based audio classification and keyword spotting on SoC FPGA

2026-02-18 · Kamil Jeziorek, Piotr Wzorek, Krzysztof Blachut, Hiroshi Nakano 외 arxiv

As the volume of data recorded by embedded edge sensors increases, particularly from neuromorphic devices producing discrete event streams, there is a growing need for hardware-aware neural architectures that enable effi…

Audio ClassificationKeyword Spotting