Multi-Objective Hardware Aware Neural Architecture Search using Hardware Cost Diversity

arXiv:2404.12403v1 Announce Type: new
Abstract: Hardware-aware Neural Architecture Search approaches (HW-NAS) automate the design of deep learning architectures, tailored specifically to a given target hardware platform. Yet, these techniques demand substantial computational resources, primarily due to the expensive process of assessing the performance of identified architectures. To alleviate this problem, a recent direction in the literature has employed representation similarity metric for efficiently evaluating architecture performance. Nonetheless, since it is inherently a single objective method, it requires multiple runs to …

abstract architecture architectures arxiv automate computational cost cs.ai cs.lg deep learning demand design diversity hardware multi-objective nas neural architecture search performance platform process resources search type