HyperAI超神経
ホームニュース最新論文チュートリアルデータセット百科事典SOTALLMモデルGPU ランキング学会
検索
サイトについて
HyperAI超神経

Neural Architecture Search On Nas Bench 201

評価指標

Accuracy (Test)
Accuracy (Val)
Search time (s)

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

モデル名
Accuracy (Test)
Accuracy (Val)
Search time (s)
Paper TitleRepository
DiNAS45.4146.6615.36Multi-conditioned Graph Diffusion for Neural Architecture Search
ParZC46.3446.37-ParZC: Parametric Zero-Cost Proxies for Efficient NAS-
AG-Net46.4246.73-Learning Where To Look -- Generative NAS is Surprisingly Efficient
GDAS41.71-28926Searching for A Robust Neural Architecture in Four GPU Hours
α-DARTS46.3446.17-$α$ DARTS Once More: Enhancing Differentiable Architecture Search by Masked Image Modeling-
EPE-NAS (N=100)38.80-20.5EPE-NAS: Efficient Performance Estimation Without Training for Neural Architecture Search
NASBOT46.37-75600Neural Architecture Search with Bayesian Optimisation and Optimal Transport
DARTS-45.1244.87-DARTS-: Robustly Stepping out of Performance Collapse Without Indicators
arch2vec46.27--Does Unsupervised Architecture Representation Learning Help Neural Architecture Search?
Shapley-NAS46.8546.57-Shapley-NAS: Discovering Operation Contribution for Neural Architecture Search
GEA46.04--Guided Evolutionary Neural Architecture Search With Efficient Performance Estimation-
iDARTS40.8940.38-iDARTS: Differentiable Architecture Search with Stochastic Implicit Gradients
PNAS +-44.75-Progressive Neural Architecture Search
DARTS-SaBN45.85--Sandwich Batch Normalization: A Drop-In Replacement for Feature Distribution Heterogeneity
DARTS (second order)16.43-29902DARTS: Differentiable Architecture Search
Λ-DARTS46.3446.37-$Λ$-DARTS: Mitigating Performance Collapse by Harmonizing Operation Selection among Cells
Local search46.38-151200Exploring the Loss Landscape in Neural Architecture Search
CATCH-meta-46.0718000CATCH: Context-based Meta Reinforcement Learning for Transferrable Architecture Search-
GAEA DARTS (ERM)46.36--Geometry-Aware Gradient Algorithms for Neural Architecture Search
NAS without training (N=10)38.33-1.7Neural Architecture Search without Training
0 of 49 row(s) selected.