HyperAI초신경

홈 뉴스 최신 연구 논문 튜토리얼 데이터셋 백과사전 SOTA LLM 모델 GPU 랭킹 컨퍼런스

한국어

HyperAI초신경

Architecture Search On Cifar 10 Image

평가 지표

FLOPS

Params

Percentage error

평가 결과

이 벤치마크에서 각 모델의 성능 결과

모델 이름	FLOPS	Params	Percentage error	Paper Title	Repository
SharpSepConvDARTS	579M	3.6M	1.98	sharpDARTS: Faster and More Accurate Differentiable Architecture Search
NAONet + c/o	-	144.6M	1.93	Neural Architecture Optimization
Soft Parameter Sharing	-	33.5	2.53	Learning Implicitly Recurrent CNNs Through Parameter Sharing
NAT-M2	291M	4.6M	2.1	Neural Architecture Transfer
arch2vec	-	3.6	2.56	Does Unsupervised Architecture Representation Learning Help Neural Architecture Search?
GATES + c/o	-	4.1M	2.58	A Generic Graph-based Neural Architecture Encoding Scheme for Predictor-based NAS
PathLevel EAS + c/o	-	14.3M	2.30	Path-Level Network Transformation for Efficient Architecture Search
EEEA-Net-C (b=5)+ CO	-	3.6	-	EEEA-Net: An Early Exit Evolutionary Neural Architecture Search
Proxyless-G + c/o	-	5.7M	2.08	ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware
MUXNet-m	200M	2.1M	2.0	MUXConv: Information Multiplexing in Convolutional Neural Networks
DARTS + c/o	-	3.4M	2.83	DARTS: Differentiable Architecture Search
NAT-M1	232M	4.3M	2.6	Neural Architecture Transfer
NAS-RL-A + c/o	-	27.6M	2.4	Neural Architecture Search with Reinforcement Learning
ENAS + c/o	-	4.6M	2.89	Efficient Neural Architecture Search via Parameter Sharing
AlphaX-1 (cutout NASNet)	-	3.59M	-	AlphaX: eXploring Neural Architectures with Deep Neural Networks and Monte Carlo Tree Search
NAT-M3	392M	6.2M	1.8	Neural Architecture Transfer
NAT-M4	468M	6.9M	1.6	Neural Architecture Transfer
CATE	-	4.1	2.46	CATE: Computation-aware Neural Architecture Encoding with Transformers
AmoebaNet-B + c/o	-	34.9M	2.13	Regularized Evolution for Image Classifier Architecture Search

0 of 19 row(s) selected.