EMEURO: A framework for generating multi-purpose accelerators via deep learning

Abstract

Approximate computing is a very promising design paradigm for crossing the CPU power wall, primarily driven by the potential to sacrifice output quality for significant gains in performance, energy, and fault tolerance. Unfortunately, existing solutions have primarily either focused on new programming models, or new hardware designs, leaving significant room between these two ends for software-based optimizations. To fill this void, additional efforts should target the compilation and runtime stages, which have a critical impact on controlling the interactions of the many approximate subcomputations to form a well-optimized application.This paper presents EMEURO, a neural-network (NN) based emulation and acceleration platform. By restructuring algorithms to have the same data flow as a NN, EMEURO is able to achieve significant speedup across several domains with minimal design effort. EMEURO uses novel NN-based approximate computing techniques, including methods for efficiently searching the high-dimension subroutine space, and fine-grain control of error during runtime. EMEURO is able to achieve 7x-109x maximum speedup over the original algorithm with 0.1%-10% approximation error. 2015 IEEE/ACM International Symposium on Code Generation and Optimization 978-1-4799-8161-8/15/$31.00 c 2015 IEEE 125 1-4799-8161-8/15/$31.00 ©2015 IEEE

Neural acceleration for GPU throughput processors

“…Recent work has explored a variety of approximation techniques that include: (a) approximate storage designs [38,39] that trades quality of data for reduced energy [38] and longer lifetime [39], (b) voltage overscaling [28,40,41], (c) loop perforation [30,42,43], (d) loop early termination [29], (e) computation substitution [6,9,29,44], (f) memoization [7,8,45], (g) limited fault recovery [42,[46][47][48][49][50], (h) precision scaling [16,51], (i) approximate circuit synthesis [19,[52][53][54][55][56][57], and (j) neural acceleration [10][11][12][13][14][15].…”

“…This characteristic of many GPU applications provides a unique opportunity to devise approximation techniques that trade small losses in the quality of results for significant gains in performance and efficiency. Among approximation techniques, neural acceleration provides significant gains for CPUs [10][11][12][13][14] and may be a good candidate for GPUs. Neural acceleration relies on an automated algorithmic transformation that converts an approximable segment of code 1 to a neural network.…”

Edge-Supported Approximate Analysis for Long Running Computations

“…Neural network-based accelerators: McAffee and Olukotun [5] developed EMEURO which is a neuralnetwork emulation and acceleration platform. With small approximation error rate, EMEURO can achieve considerable speedup in various applications within the image processing domain.…”

Optimization of program execution using computational significance

“…These frameworks delegate the control of approximate code execution to the programmer. Emeuro [60] efficiently breaks down an application into subroutines of varying granularity and automatically generates approximate alternatives for said subroutines through the use of Artificial Neural Networks (ANNs). At execution time, an intelligent runtime system explores the high-dimension subroutine space and generates a graph of computations which comprises nodes that are either accurate versions of the subroutines approximate ones through the use of ANNs.…”