SOTAVerified

Neural Architecture Search

Neural architecture search (NAS) is a technique for automating the design of artificial neural networks (ANN), a widely used model in the field of machine learning. NAS essentially takes the process of a human manually tweaking a neural network and learning what works well, and automates this task to discover more complex architectures.

Image Credit : NAS with Reinforcement Learning

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 151–200 of 1915 papers

TitleStatusHype
KNAS: Green Neural Architecture SearchCode1
EH-DNAS: End-to-End Hardware-aware Differentiable Neural Architecture SearchCode1
DU-DARTS: Decreasing the Uncertainty of Differentiable Architecture SearchCode1
Pruning Self-attentions into Convolutional Layers in Single PathCode1
Rethinking Dilated Convolution for Real-time Semantic SegmentationCode1
deepstruct -- linking deep learning and graph theoryCode1
NAS-Bench-x11 and the Power of Learning CurvesCode1
One Proxy Device Is Enough for Hardware-Aware Neural Architecture SearchCode1
MCUNetV2: Memory-Efficient Patch-based Inference for Tiny Deep LearningCode1
NAS-FCOS: Efficient Search for Object Detection ArchitecturesCode1
Grafting Transformer on Automatically Designed Convolutional Neural Network for Hyperspectral Image ClassificationCode1
NAS-HPO-Bench-II: A Benchmark Dataset on Joint Optimization of Convolutional Neural Network Architecture and Training HyperparametersCode1
NAS-Bench-360: Benchmarking Neural Architecture Search on Diverse TasksCode1
Dataset Condensation with Distribution MatchingCode1
Multi-objective Optimization by Learning Space PartitionsCode1
NASViT: Neural Architecture Search for Efficient Vision Transformers with Gradient Conflict aware Supernet TrainingCode1
AutoInit: Analytic Signal-Preserving Weight Initialization for Neural NetworksCode1
SpeechNAS: Towards Better Trade-off between Latency and Accuracy for Large-Scale Speaker VerificationCode1
RobustART: Benchmarking Robustness on Architecture Design and Training TechniquesCode1
ReconfigISP: Reconfigurable Camera Image Processing PipelineCode1
RepNAS: Searching for Efficient Re-parameterizing BlocksCode1
NAS-OoD: Neural Architecture Search for Out-of-Distribution GeneralizationCode1
Searching for Efficient Multi-Stage Vision TransformersCode1
Understanding and Accelerating Neural Architecture Search with Training-Free and Theory-Grounded MetricsCode1
Pooling Architecture Search for Graph ClassificationCode1
PocketNet: Extreme Lightweight Face Recognition Network using Neural Architecture Search and Multi-Step Knowledge DistillationCode1
Pi-NAS: Improving Neural Architecture Search by Reducing Supernet Training Consistency ShiftCode1
BN-NAS: Neural Architecture Search with Batch NormalizationCode1
FOX-NAS: Fast, On-device and Explainable Neural Architecture SearchCode1
EEEA-Net: An Early Exit Evolutionary Neural Architecture SearchCode1
Rethinking Architecture Selection in Differentiable NASCode1
Accelerating Evolutionary Neural Architecture Search via Multi-Fidelity EvaluationCode1
Generic Neural Architecture Search via RegressionCode1
AdvRush: Searching for Adversarially Robust Neural ArchitecturesCode1
GLiT: Neural Architecture Search for Global and Local Image TransformerCode1
Integrating Large Circular Kernels into CNNs through Neural Architecture SearchCode1
Rapid Neural Architecture Search by Learning to Generate Graphs from DatasetsCode1
BiX-NAS: Searching Efficient Bi-directional Architecture for Medical Image SegmentationCode1
ViTAS: Vision Transformer Architecture SearchCode1
iDARTS: Differentiable Architecture Search with Stochastic Implicit GradientsCode1
Removing Raindrops and Rain Streaks in One GoCode1
HELP: Hardware-Adaptive Efficient Latency Prediction for NAS via Meta-LearningCode1
HR-NAS: Searching Efficient High-Resolution Neural Architectures with Lightweight TransformersCode1
Accelerating Neural Architecture Search via Proxy DataCode1
FEAR: A Simple Lightweight Method to Rank ArchitecturesCode1
Neural Architecture Search via Bregman IterationsCode1
Differentiable Architecture Search for Reinforcement LearningCode1
FuSeConv: Fully Separable Convolutions for Fast Inference on Systolic ArraysCode1
TransNAS-Bench-101: Improving Transferability and Generalizability of Cross-Task Neural Architecture SearchCode1
AutoReCon: Neural Architecture Search-based Reconstruction for Data-free CompressionCode1
Show:102550
← PrevPage 4 of 39Next →
All datasetsImageNetNAS-Bench-201, ImageNet-16-120CIFAR-10NAS-Bench-201, CIFAR-100NAS-Bench-201, CIFAR-10CIFAR-10 Image ClassificationCIFAR-100NATS-Bench Topology, CIFAR-10NATS-Bench Topology, CIFAR-100NATS-Bench Topology, ImageNet16-120Food-101NAS-Bench-101

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1SPOS (ProxylessNAS (GPU) latency)Accuracy75.3—Unverified
2SPOS (FBNet-C latency)Accuracy75.1—Unverified
3SPOS (block search + channel search)Accuracy74.7—Unverified
4MUXNet-xsTop-1 Error Rate33.3—Unverified
5FBNetV2-F1Top-1 Error Rate31.7—Unverified
6LayerNAS-60MTop-1 Error Rate31—Unverified
7NASGEPTop-1 Error Rate29.51—Unverified
8MUXNet-sTop-1 Error Rate28.4—Unverified
9NN-MASS-ATop-1 Error Rate27.1—Unverified
10FBNetV2-F3Top-1 Error Rate26.8—Unverified
#ModelMetricClaimedVerifiedStatus
1CR-LSOAccuracy (Test)46.98—Unverified
2Shapley-NASAccuracy (Test)46.85—Unverified
3β-RDARTS-L2Accuracy (Test)46.71—Unverified
4β-SDARTS-RSAccuracy (Test)46.71—Unverified
5ASE-NAS+Accuracy (Val)46.66—Unverified
6NARAccuracy (Test)46.66—Unverified
7BaLeNAS-TFAccuracy (Test)46.54—Unverified
8AG-NetAccuracy (Test)46.42—Unverified
9Local searchAccuracy (Test)46.38—Unverified
10NASBOTAccuracy (Test)46.37—Unverified
#ModelMetricClaimedVerifiedStatus
1Balanced MixtureAccuracy (% )91.55—Unverified
2GDASTop-1 Error Rate3.4—Unverified
3Bonsai-NetTop-1 Error Rate3.35—Unverified
4Net2 (2)Top-1 Error Rate3.3—Unverified
5μDARTSTop-1 Error Rate3.28—Unverified
6NN-MASS- CIFAR-CTop-1 Error Rate3.18—Unverified
7NN-MASS- CIFAR-ATop-1 Error Rate3—Unverified
8DARTS (first order)Top-1 Error Rate3—Unverified
9NASGEPTop-1 Error Rate2.82—Unverified
10AlphaX-1 (cutout NASNet)Top-1 Error Rate2.82—Unverified