NADER uses a multi-agent LLM team with a graph-based block representation and a reflection memory to iteratively propose and test modified neural architectures, claiming gains beyond NAS-Bench-201's optimum on CIFAR and ImageNet16-120.
Random search for hyper-parameter optimization
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
NADER: Neural Architecture Design via Multi-Agent Collaboration
NADER uses a multi-agent LLM team with a graph-based block representation and a reflection memory to iteratively propose and test modified neural architectures, claiming gains beyond NAS-Bench-201's optimum on CIFAR and ImageNet16-120.