Hardware-Guided Symbiotic Training for Compact, Accurate, yet Execution-Efficient LSTM
January 30, 2019 ยท Declared Dead ยท ๐ arXiv.org
"No code URL or promise found in abstract"
Evidence collected by the PWNC Scanner
Authors
Hongxu Yin, Guoyang Chen, Yingmin Li, Shuai Che, Weifeng Zhang, Niraj K. Jha
arXiv ID
1901.10997
Category
cs.NE: Neural & Evolutionary
Citations
10
Venue
arXiv.org
Last Checked
4 months ago
Abstract
Many long short-term memory (LSTM) applications need fast yet compact models. Neural network compression approaches, such as the grow-and-prune paradigm, have proved to be promising for cutting down network complexity by skipping insignificant weights. However, current compression strategies are mostly hardware-agnostic and network complexity reduction does not always translate into execution efficiency. In this work, we propose a hardware-guided symbiotic training methodology for compact, accurate, yet execution-efficient inference models. It is based on our observation that hardware may introduce substantial non-monotonic behavior, which we call the latency hysteresis effect, when evaluating network size vs. inference latency. This observation raises question about the mainstream smaller-dimension-is-better compression strategy, which often leads to a sub-optimal model architecture. By leveraging the hardware-impacted hysteresis effect and sparsity, we are able to achieve the symbiosis of model compactness and accuracy with execution efficiency, thus reducing LSTM latency while increasing its accuracy. We have evaluated our algorithms on language modeling and speech recognition applications. Relative to the traditional stacked LSTM architecture obtained for the Penn Treebank dataset, we reduce the number of parameters by 18.0x (30.5x) and measured run-time latency by up to 2.4x (5.2x) on Nvidia GPUs (Intel Xeon CPUs) without any accuracy degradation. For the DeepSpeech2 architecture obtained for the AN4 dataset, we reduce the number of parameters by 7.0x (19.4x), word error rate from 12.9% to 9.9% (10.4%), and measured run-time latency by up to 1.7x (2.4x) on Nvidia GPUs (Intel Xeon CPUs). Thus, our method yields compact, accurate, yet execution-efficient inference models.
Community Contributions
Found the code? Know the venue? Think something is wrong? Let us know!
๐ Similar Papers
In the same crypt โ Neural & Evolutionary
๐ฎ
๐ฎ
The Ethereal
R.I.P.
๐ป
Ghosted
Deep Learning using Rectified Linear Units (ReLU)
R.I.P.
๐ป
Ghosted
Generative Adversarial Text to Image Synthesis
R.I.P.
๐ป
Ghosted
Regularized Evolution for Image Classifier Architecture Search
R.I.P.
๐ป
Ghosted
Temporal Ensembling for Semi-Supervised Learning
๐
๐
Old Age
Learning Structured Sparsity in Deep Neural Networks
Died the same way โ ๐ป Ghosted
R.I.P.
๐ป
Ghosted
Federated Learning: Strategies for Improving Communication Efficiency
R.I.P.
๐ป
Ghosted
In-Datacenter Performance Analysis of a Tensor Processing Unit
R.I.P.
๐ป
Ghosted
Deep Convolutional Neural Networks for Computer-Aided Detection: CNN Architectures, Dataset Characteristics and Transfer Learning
R.I.P.
๐ป
Ghosted