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

Showing 16761700 of 1915 papers

TitleStatusHype
Fair DARTS: Eliminating Unfair Advantages in Differentiable Architecture SearchCode0
Boosting Order-Preserving and Transferability for Neural Architecture Search: a Joint Architecture Refined Search and Fine-tuning ApproachCode0
ShiftNAS: Improving One-shot NAS via Probability ShiftCode0
MFAS: Multimodal Fusion Architecture SearchCode0
Bonsai-Net: One-Shot Neural Architecture Search via Differentiable PrunersCode0
Adaptive Search-and-Training for Robust and Efficient Network PruningCode0
BOAH: A Tool Suite for Multi-Fidelity Bayesian Optimization & Analysis of HyperparametersCode0
Shortest Edit Path Crossover: A Theory-driven Solution to the Permutation Problem in Evolutionary Neural Architecture SearchCode0
BLOX: Macro Neural Architecture Search Benchmark and AlgorithmsCode0
Exploring Unexplored Tensor Network Decompositions for Convolutional Neural NetworksCode0
Exploring Robust Architectures for Deep Artificial Neural NetworksCode0
Exploring Randomly Wired Neural Networks for Image RecognitionCode0
Tree ensemble kernels for Bayesian optimization with known constraints over mixed-feature spacesCode0
Blending Diverse Physical Priors with Neural NetworksCode0
Ps and Qs: Quantization-aware pruning for efficient low latency neural network inferenceCode0
Shrink-Perturb Improves Architecture Mixing during Population Based Training for Neural Architecture SearchCode0
Experiments on Properties of Hidden Structures of Sparse Neural NetworksCode0
Mixture-of-Supernets: Improving Weight-Sharing Supernet Training with Architecture-Routed Mixture-of-ExpertsCode0
Evolving Robust Neural Architectures to Defend from Adversarial AttacksCode0
Evolving Neural Networks through Augmenting TopologiesCode0
MobileDets: Searching for Object Detection Architectures for Mobile AcceleratorsCode0
Evolving Neural Architecture Using One Shot ModelCode0
Pseudo-Inverted Bottleneck Convolution for DARTS Search SpaceCode0
Evolutionary NAS with Gene Expression Programming of Cellular EncodingCode0
TreeGrad: Transferring Tree Ensembles to Neural NetworksCode0
Show:102550
← PrevPage 68 of 77Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1SPOS (ProxylessNAS (GPU) latency)Accuracy75.3Unverified
2SPOS (FBNet-C latency)Accuracy75.1Unverified
3SPOS (block search + channel search)Accuracy74.7Unverified
4MUXNet-xsTop-1 Error Rate33.3Unverified
5FBNetV2-F1Top-1 Error Rate31.7Unverified
6LayerNAS-60MTop-1 Error Rate31Unverified
7NASGEPTop-1 Error Rate29.51Unverified
8MUXNet-sTop-1 Error Rate28.4Unverified
9NN-MASS-ATop-1 Error Rate27.1Unverified
10FBNetV2-F3Top-1 Error Rate26.8Unverified
#ModelMetricClaimedVerifiedStatus
1CR-LSOAccuracy (Test)46.98Unverified
2Shapley-NASAccuracy (Test)46.85Unverified
3β-RDARTS-L2Accuracy (Test)46.71Unverified
4β-SDARTS-RSAccuracy (Test)46.71Unverified
5ASE-NAS+Accuracy (Val)46.66Unverified
6NARAccuracy (Test)46.66Unverified
7BaLeNAS-TFAccuracy (Test)46.54Unverified
8AG-NetAccuracy (Test)46.42Unverified
9Local searchAccuracy (Test)46.38Unverified
10NASBOTAccuracy (Test)46.37Unverified
#ModelMetricClaimedVerifiedStatus
1Balanced MixtureAccuracy (% )91.55Unverified
2GDASTop-1 Error Rate3.4Unverified
3Bonsai-NetTop-1 Error Rate3.35Unverified
4Net2 (2)Top-1 Error Rate3.3Unverified
5μDARTSTop-1 Error Rate3.28Unverified
6NN-MASS- CIFAR-CTop-1 Error Rate3.18Unverified
7NN-MASS- CIFAR-ATop-1 Error Rate3Unverified
8DARTS (first order)Top-1 Error Rate3Unverified
9NASGEPTop-1 Error Rate2.82Unverified
10AlphaX-1 (cutout NASNet)Top-1 Error Rate2.82Unverified