High performance Block-Based Neural Network Model by pipelined parallel communication

Lee, Kundo; Hamagami, Tomoki

doi:10.1541/ieejeiss.139.1059

Cited by 1 publication

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we introduce a new approach in reservoir implementation using the Asynchronous Block-Based Neural Network model to overcome the issues in implementing RC on dedicated hardware. ABBNN [14] is a high-performance model of Block-Based Neural Network (BBNN) [15]. This paper is organized as follows: Section 2 introduces Reservoir Computing and Echo State Networks, which provide a novel approach to supervised training of RNN.…”

Section: Introductionmentioning

confidence: 99%

Reservoir Computing for Scalable Hardware with Block‐Based Neural Network

Lee

Hamagami

2021

IEEJ Transactions Elec Engng

Self Cite

View full text Add to dashboard Cite

This paper presents a reservoir computing technique using Asynchronous Block-Based Neural Networks (ABBNN). Echo State Networks and Liquid State Machines introduced a new paradigm in artificial neural networks (ANN), known as Reservoir Computing (RC). A recurrent neural network (RNN) in Reservoir Computing is generated randomly, and only a readout is trained. The reservoir computing greatly facilitated the practical RNN application and outperformed classical RNN in many tasks. ABBNN is an extended version of the classical block-based neural network model. ABBNN, an evolvable neural network model, provides a model-free estimation of nonlinear dynamical systems. To propose a hardware-aware model of reservoir computing and high-speed training, we introduce Block-Based Reservoir Computing (BBRC ). BBRC provides a flexible architecture. The architecture based on the basic blocks provides a regularity that supports scalable hardware. In contrast, BBRC also supports randomness based on the internal configuration. Both are significant features in achieving an optimal hardware reservoir computer.

show abstract

Section: Introductionmentioning

confidence: 99%