Neural Architecture Search On Cifar 10

평가 지표

Parameters

Search Time (GPU days)

Top-1 Error Rate

평가 결과

이 벤치마크에서 각 모델의 성능 결과

				Paper Title
GDAS	-	0.21	3.4%	Searching for A Robust Neural Architecture in Four GPU Hours
Bonsai-Net	2.9M	0.10	3.35%	Bonsai-Net: One-Shot Neural Architecture Search via Differentiable Pruners
Net2 (2)	-	-	3.3%	Efficacy of Neural Prediction-Based Zero-Shot NAS
μDARTS	-	0.1	3.277%	$μ$DARTS: Model Uncertainty-Aware Differentiable Architecture Search
NN-MASS- CIFAR-C	3.82M	0	3.18%	How does topology of neural architectures impact gradient propagation and model performance?
NN-MASS- CIFAR-A	5.02M	0	3.0%	How does topology of neural architectures impact gradient propagation and model performance?
DARTS (first order)	3.3	1.5	3%	DARTS: Differentiable Architecture Search
NASGEP	-	1	2.82%	Optimizing Neural Architecture Search using Limited GPU Time in a Dynamic Search Space: A Gene Expression Programming Approach
AlphaX-1 (cutout NASNet)	-	224	2.82%	AlphaX: eXploring Neural Architectures with Deep Neural Networks and Monte Carlo Tree Search
DARTS (second order)	3.3	4	2.76%	DARTS: Differentiable Architecture Search
SETN (T=1K) + CutOut	-	1.8	2.69%	One-Shot Neural Architecture Search via Self-Evaluated Template Network
DARTS-PRIME	3.7M	0.5	2.62%	DARTS-PRIME: Regularization and Scheduling Improve Constrained Optimization in Differentiable NAS
NAT-M1	4.3M	1.0	2.6%	Neural Architecture Transfer
PC-DARTS	3.6M	0.1	2.57%	PC-DARTS: Partial Channel Connections for Memory-Efficient Architecture Search
arch2vec	3.6M	10.5	2.56%	Does Unsupervised Architecture Representation Learning Help Neural Architecture Search?
FairDARTS-a	2.8M	0.25	2.54%	Fair DARTS: Eliminating Unfair Advantages in Differentiable Architecture Search
MSR-DARTS	4.0M	0.3	2.54%	MSR-DARTS: Minimum Stable Rank of Differentiable Architecture Search
Soft Parameter Sharing	-	0.7	2.53%	Learning Implicitly Recurrent CNNs Through Parameter Sharing
β-DARTS	-	-	2.53%	$β$-DARTS: Beta-Decay Regularization for Differentiable Architecture Search
TNASP	3.7M	0.3	2.52%	TNASP: A Transformer-based NAS Predictor with a Self-evolution Framework

0 of 41 row(s) selected.

Command Palette

Neural Architecture Search On Cifar 10

평가 지표

평가 결과