Neural Architecture Search (NAS)

Share

Neural Architecture Search (NAS) is the process of automating network architecture engineering or finding the optimal design of artificial neural networks. The methods for NAS include three components: the search space, search strategy, and performance estimation strategy. It is a subfield of automated machine learning (AutoML). AutoNAC has a NAS component that redesigns a given trained model’s architecture to optimally improve its inference performance (throughput, latency, memory, etc.) for specific target hardware while preserving its baseline accuracy.

 

Filter terms by

Related resources

deci-winter-release-2023-blog-featured-5
Algorithms
deci-mlperf-cv-cover
Research
Image depicting a GPU
Research