Recurrent Neural Networks: An Embedded Computing Perspective

TL;DR

This paper reviews the challenges and solutions for implementing Recurrent Neural Networks on embedded devices, emphasizing optimization techniques, current limitations, and future research directions for efficient and flexible RNN deployment.

Contribution

It provides a comprehensive review of RNN implementations on embedded systems, highlighting optimization strategies and identifying gaps in flexibility and efficiency.

Findings

Algorithmic optimizations are crucial for efficiency.

Memory access reduction improves RNN performance.

High performance is prioritized over flexibility in current implementations.

Abstract

Recurrent Neural Networks (RNNs) are a class of machine learning algorithms used for applications with time-series and sequential data. Recently, there has been a strong interest in executing RNNs on embedded devices. However, difficulties have arisen because RNN requires high computational capability and a large memory space. In this paper, we review existing implementations of RNN models on embedded platforms and discuss the methods adopted to overcome the limitations of embedded systems. We will define the objectives of mapping RNN algorithms on embedded platforms and the challenges facing their realization. Then, we explain the components of RNN models from an implementation perspective. We also discuss the optimizations applied to RNNs to run efficiently on embedded platforms. Finally, we compare the defined objectives with the implementations and highlight some open research questions and aspects currently not addressed for embedded RNNs. Overall, applying algorithmic optimizations to RNN models and decreasing the memory access overhead is vital to obtain high efficiency. To further increase the implementation efficiency, we point up the more promising optimizations that could be applied in future research. Additionally, this article observes that high performance has been targeted by many implementations, while flexibility has, as yet, been attempted less often. Thus, the article provides some guidelines for RNN hardware designers to support flexibility in a better manner.