SeqPoint: Identifying Representative Iterations of Sequence-based Neural Networks

07/20/2020
by   Suchita Pati, et al.
0

The ubiquity of deep neural networks (DNNs) continues to rise, making them a crucial application class for hardware optimizations. However, detailed profiling and characterization of DNN training remains difficult as these applications often run for hours to days on real hardware. Prior works exploit the iterative nature of DNNs to profile a few training iterations. While such a strategy is sound for networks like convolutional neural networks (CNNs), where the nature of the computation is largely input independent, we observe in this work that this approach is sub-optimal for sequence-based neural networks (SQNNs) such as recurrent neural networks (RNNs). The amount and nature of computations in SQNNs can vary for each input, resulting in heterogeneity across iterations. Thus, arbitrarily selecting a few iterations is insufficient to accurately summarize the behavior of the entire training run. To tackle this challenge, we carefully study the factors that impact SQNN training iterations and identify input sequence length as the key determining factor for variations across iterations. We then use this observation to characterize all iterations of an SQNN training run (requiring no profiling or simulation of the application) and select representative iterations, which we term SeqPoints. We analyze two state-of-the-art SQNNs, DeepSpeech2 and Google's Neural Machine Translation (GNMT), and show that SeqPoints can represent their entire training runs accurately, resulting in geomean errors of only 0.11 respectively, when projecting overall runtime and 0.13 projecting speedups due to architectural changes. This high accuracy is achieved while reducing the time needed for profiling by 345x and 214x for the two networks compared to full training runs. As a result, SeqPoint can enable analysis of SQNN training runs in mere minutes instead of hours or days.

READ FULL TEXT
research
11/09/2019

Optimizing Deep Learning Inference on Embedded Systems Through Adaptive Model Selection

Deep neural networks ( DNNs ) are becoming a key enabling technology for...
research
09/29/2021

On the Provable Generalization of Recurrent Neural Networks

Recurrent Neural Network (RNN) is a fundamental structure in deep learni...
research
02/14/2022

Saving RNN Computations with a Neuron-Level Fuzzy Memoization Scheme

Recurrent Neural Networks (RNNs) are a key technology for applications s...
research
08/28/2019

Distributed Deep Learning for Precipitation Nowcasting

Effective training of Deep Neural Networks requires massive amounts of d...
research
01/23/2011

Building a Chaotic Proved Neural Network

Chaotic neural networks have received a great deal of attention these la...
research
03/03/2020

Selectivity considered harmful: evaluating the causal impact of class selectivity in DNNs

Class selectivity, typically defined as how different a neuron's respons...
research
06/16/2023

Representation and decomposition of functions in DAG-DNNs and structural network pruning

The conclusions provided by deep neural networks (DNNs) must be carefull...

Please sign up or login with your details

Forgot password? Click here to reset