FOX-NAS: Fast, On-device and Explainable Neural Architecture Search

Liu, Chia-Hsiang; Han, Yu-Shin; Sung, Yuan-Yao; Lee, Yi; Chiang, Hung-Yueh; Wu, Kai–Chiang

doi:10.48550/arxiv.2108.08189

Cited by 1 publication

(1 citation statement)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A promising approach is to train a regression model on a dataset of architecture and latency pairs, collected from the target device [34,32,40]. The regression model can predict the latency of unseen architectures and removes the need to measure the latency of every architecture during the NAS process; significantly decreases the amount of time spent on acquiring latency measurements.…”

Section: Related Workmentioning

confidence: 99%

MAPLE: Microprocessor A Priori for Latency Estimation

Abbasi¹,

Wong²,

Shafiee³

2021

Preprint

View full text Add to dashboard Cite

Modern deep neural networks must demonstrate state-of-the-art accuracy while exhibiting low latency and energy consumption. As such, neural architecture search (NAS) algorithms take these two constraints into account when generating a new architecture. However, efficiency metrics such as latency are typically hardware dependent requiring the NAS algorithm to either measure or predict the architecture latency. Measuring the latency of every evaluated architecture adds a significant amount of time to the NAS process. Here we propose Microprocessor A Priori for Latency Estimation MAPLE that does not rely on transfer learning or domain adaptation but instead generalizes to new hardware by incorporating a prior hardware characteristics during training. MAPLE takes advantage of a novel quantitative strategy to characterize the underlying microprocessor by measuring relevant hardware performance metrics, yielding a fine-grained and expressive hardware descriptor. Moreover, the proposed MAPLE benefits from the tightly coupled I/O between the CPU and GPU and their dependency to predict DNN latency on GPUs while measuring microprocessor performance hardware counters from the CPU feeding the GPU hardware. Through this quantitative strategy as the hardware descriptor, MAPLE can generalize to new hardware via a few shot adaptation strategy where with as few as 3 samples it exhibits a 3% improvement over state-of-the-art methods requiring as much as 10 samples. Experimental results showed that, increasing the few shot adaptation samples to 10 improves the accuracy significantly over the state-of-the-art methods by 12%. Furthermore, it was demonstrated that MAPLE exhibiting 8-10% better accuracy, on average, compared to relevant baselines at any number of adaptation samples. The proposed method provides a versatile and practical latency prediction methodology inferring DNN run-time on multiple hardware devices while not imposing any significant overhead for sample collection.Preprint. Under review.

show abstract

Section: Related Workmentioning

confidence: 99%