Neural Architecture Search On Nas Bench 201 2

Metrics

Accuracy (Test)

Accuracy (Val)

Search time (s)

Results

Performance results of various models on this benchmark

Model Name	Accuracy (Test)	Accuracy (Val)	Search time (s)	Paper Title	Repository
DARTS-V1	15.61	15.03	10890	-	-
KNAS (k=40)	71.05	-	-	KNAS: Green Neural Architecture Search
DSNAS	-	31.01	-	DSNAS: Direct Neural Architecture Search without Parameter Retraining
SETN	56.87	59.05	31010	One-Shot Neural Architecture Search via Self-Evaluated Template Network
Shapley-NAS	73.51	73.49	-	Shapley-NAS: Discovering Operation Contribution for Neural Architecture Search
IS-DARTS	73.51	73.49	-	IS-DARTS: Stabilizing DARTS through Precise Measurement on Candidate Importance
RF-DARTS	72.94	72.95	-	Differentiable Architecture Search with Random Features	-
α-DARTS	73.16	73.21	-	$α$ DARTS Once More: Enhancing Differentiable Architecture Search by Masked Image Modeling	-
SNAS	69.34	69.69	-	SNAS: Stochastic Neural Architecture Search
DARTS-V2	15.61	15.03	29902	-	-
iDARTS	70.83	70.57	-	iDARTS: Differentiable Architecture Search with Stochastic Implicit Gradients
DrNAS	73.51	73.49	-	DrNAS: Dirichlet Neural Architecture Search
GAEA DARTS (ERM)	73.43	-	-	Geometry-Aware Gradient Algorithms for Neural Architecture Search
ENAS	15.61	15.03	13315	Efficient Neural Architecture Search via Parameters Sharing	-
BaLeNAS-TF	72.95	72.67	-	BaLeNAS: Differentiable Architecture Search via the Bayesian Learning Rule	-
GenNAS	72.56	-	1080	Generic Neural Architecture Search via Regression
DiNAS	73.51	73.49	15.36	Multi-conditioned Graph Diffusion for Neural Architecture Search
arch2vec	73.37	73.35	-	Does Unsupervised Architecture Representation Learning Help Neural Architecture Search?
CR-LSO	73.47	73.44	-	CR-LSO: Convex Neural Architecture Optimization in the Latent Space of Graph Variational Autoencoder with Input Convex Neural Networks
DARTS-	71.53	71.36	-	DARTS-: Robustly Stepping out of Performance Collapse Without Indicators

0 of 40 row(s) selected.