DeepAI AI Chat
Log In Sign Up

Hardware-oriented Approximation of Convolutional Neural Networks

by   Philipp Gysel, et al.
University of California-Davis

High computational complexity hinders the widespread usage of Convolutional Neural Networks (CNNs), especially in mobile devices. Hardware accelerators are arguably the most promising approach for reducing both execution time and power consumption. One of the most important steps in accelerator development is hardware-oriented model approximation. In this paper we present Ristretto, a model approximation framework that analyzes a given CNN with respect to numerical resolution used in representing weights and outputs of convolutional and fully connected layers. Ristretto can condense models by using fixed point arithmetic and representation instead of floating point. Moreover, Ristretto fine-tunes the resulting fixed point network. Given a maximum error tolerance of 1 The code for Ristretto is available.


page 1

page 2

page 3

page 4


Ristretto: Hardware-Oriented Approximation of Convolutional Neural Networks

Convolutional neural networks (CNN) have achieved major breakthroughs in...

A scalable and efficient convolutional neural network accelerator using HLS for a System on Chip design

This paper presents a configurable Convolutional Neural Network Accelera...

CORDIC-based Architecture for Powering Computation in Fixed-Point Arithmetic

We present a fixed point architecture (source VHDL code is provided) for...

FPGA Implementation of Convolutional Neural Networks with Fixed-Point Calculations

Neural network-based methods for image processing are becoming widely us...

Method for Hybrid Precision Convolutional Neural Network Representation

This invention addresses fixed-point representations of convolutional ne...

Stochastic Rounding: Algorithms and Hardware Accelerator

Algorithms and a hardware accelerator for performing stochastic rounding...

Self-Compressing Neural Networks

This work focuses on reducing neural network size, which is a major driv...