Hardware Implementation of Hyperbolic Tangent Function using Catmull-Rom Spline Interpolation

TL;DR

This paper proposes a hardware implementation of the hyperbolic tangent function using Catmull-Rom spline interpolation, achieving high accuracy with reduced logic area for neural network accelerators.

Contribution

It introduces a novel method for implementing tanh in hardware using spline interpolation, improving efficiency over existing approaches.

Findings

Achieves accurate tanh approximation with smaller logic area.

Demonstrates suitability for neural network hardware accelerators.

Provides performance benchmarks showing efficiency gains.

Abstract

Deep neural networks yield the state of the art results in many computer vision and human machine interface tasks such as object recognition, speech recognition etc. Since, these networks are computationally expensive, customized accelerators are designed for achieving the required performance at lower cost and power. One of the key building blocks of these neural networks is non-linear activation function such as sigmoid, hyperbolic tangent (tanh), and ReLU. A low complexity accurate hardware implementation of the activation function is required to meet the performance and area targets of the neural network accelerators. This paper presents an implementation of tanh function using the Catmull-Rom spline interpolation. State of the art results are achieved using this method with comparatively smaller logic area.