Title
Design Automation for Efficient Deep Learning Computing.
Abstract
Efficient deep learning computing requires algorithm and hardware co-design to enable specialization: we usually need to change the algorithm to reduce memory footprint and improve energy efficiency. However, the extra degree of freedom from the algorithm makes the design space much larger: itu0027s not only about designing the hardware but also about how to tweak the algorithm to best fit the hardware. Human engineers can hardly exhaust the design space by heuristics. Itu0027s labor consuming and sub-optimal. We propose design automation techniques for efficient neural networks. We investigate automatically designing specialized fast models, auto channel pruning, and auto mixed-precision quantization. We demonstrate such learning-based, automated design achieves superior performance and efficiency than rule-based human design. Moreover, we shorten the design cycle by 200x than previous work, so that we can afford to design specialized neural network models for different hardware platforms.
Year
Venue
DocType
2019
arXiv: Learning
Journal
Citations 
PageRank 
References 
2
0.36
0
Authors
7
Name
Order
Citations
PageRank
Song Han1210279.81
Han Cai222310.39
Ligen Zhu3835.19
Lin, Ji4798.18
Kuan Wang521.71
Zhijian Liu6599.80
Yujun Lin720.36