ZAPS pipeline enhances Neural Architecture Search by combining proxy signals and topology

Researchers have developed ZAPS, a novel four-stage pipeline designed to improve Neural Architecture Search (NAS) by efficiently combining proxy signals with architectural topology. This method addresses the limitations of existing zero-cost proxies, which are often noisy and correlated, by selecting a non-redundant subset of proxies and re-evaluating them iteratively. ZAPS utilizes a hybrid k-means strategy for initial seeding and an XGBoost ensemble with a UCB acquisition function for ranking candidates, outperforming several baseline methods on NAS-Bench-201 for CIFAR-10 and CIFAR-100. AI

IMPACT Improves efficiency and accuracy in designing neural networks, potentially accelerating AI model development.

RANK_REASON The item is a research paper detailing a new method for Neural Architecture Search. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• banana
• CIFAR-10
• CIFAR-100
• k-means clustering
• local search
• NAS-Bench-101
• NAS-Bench-201
• Neural Architecture Search
• ProxyFit
• random search
• Rea
• TPE
• University of California, Berkeley
• XGBoost

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →