# GreedyNAS: Towards Fast One-Shot NAS with Greedy Supernet

arXiv: https://arxiv.org/abs/2003.11236 · HTML: https://mlsanity.com/paper/2003.11236/

## Key facts

- Cited by 69 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Mar 2020.

## Built on by (most cited first, 30 of 69)

- [LocalMamba: Visual State Space Model with Windowed Selective Scan](https://mlsanity.com/paper/2403.09338/) (2024)
- [Knowledge Distillation from A Stronger Teacher](https://mlsanity.com/paper/2205.10536/) (2022)
- [SimMatch: Semi-supervised Learning with Similarity Matching](https://mlsanity.com/paper/2203.06915/) (2022)
- [Distilling Object Detectors via Decoupled Features](https://mlsanity.com/paper/2103.14475/) (2021)
- [Reformulating HOI Detection as Adaptive Set Prediction](https://mlsanity.com/paper/2103.05983/) (2021)
- [HW-NAS-Bench:Hardware-Aware Neural Architecture Search Benchmark](https://mlsanity.com/paper/2103.10584/) (2021)
- [Weakly Supervised Contrastive Learning](https://mlsanity.com/paper/2110.04770/) (2021)
- [Can GPT-4 Perform Neural Architecture Search?](https://mlsanity.com/paper/2304.10970/) (2023)
- [Manifold Regularized Dynamic Network Pruning](https://mlsanity.com/paper/2103.05861/) (2021)
- [Searching the Search Space of Vision Transformer](https://mlsanity.com/paper/2111.14725/) (2021)
- [AutoSNN: Towards Energy-Efficient Spiking Neural Networks](https://mlsanity.com/paper/2201.12738/) (2022)
- [LightViT: Towards Light-Weight Convolution-Free Vision Transformers](https://mlsanity.com/paper/2207.05557/) (2022)
- [HR-NAS: Searching Efficient High-Resolution Neural Architectures with Lightweight Transformers](https://mlsanity.com/paper/2106.06560/) (2021)
- [KNAS: Green Neural Architecture Search](https://mlsanity.com/paper/2111.13293/) (2021)
- [Prioritized Architecture Sampling with Monto-Carlo Tree Search](https://mlsanity.com/paper/2103.11922/) (2021)
- [ViTAS: Vision Transformer Architecture Search](https://mlsanity.com/paper/2106.13700/) (2021)
- [Towards Improving the Consistency, Efficiency, and Flexibility of Differentiable Neural Architecture Search](https://mlsanity.com/paper/2101.11342/) (2021)
- [Locally Free Weight Sharing for Network Width Search](https://mlsanity.com/paper/2102.05258/) (2021)
- [BCNet: Searching for Network Width with Bilaterally Coupled Network](https://mlsanity.com/paper/2105.10533/) (2021)
- [Masked Distillation with Receptive Tokens](https://mlsanity.com/paper/2205.14589/) (2022)
- [K-shot NAS: Learnable Weight-Sharing for NAS with K-shot Supernets](https://mlsanity.com/paper/2106.06442/) (2021)
- [CATE: Computation-aware Neural Architecture Encoding with Transformers](https://mlsanity.com/paper/2102.07108/) (2021)
- [BN-NAS: Neural Architecture Search with Batch Normalization](https://mlsanity.com/paper/2108.07375/) (2021)
- [NAS-Bench-x11 and the Power of Learning Curves](https://mlsanity.com/paper/2111.03602/) (2021)
- [Searching by Generating: Flexible and Efficient One-Shot NAS with Architecture Generator](https://mlsanity.com/paper/2103.07289/) (2021)
- [Data-Free Knowledge Distillation with Soft Targeted Transfer Set Synthesis](https://mlsanity.com/paper/2104.04868/) (2021)
- [HyperSegNAS: Bridging One-Shot Neural Architecture Search with 3D Medical Image Segmentation using HyperNet](https://mlsanity.com/paper/2112.10652/) (2021)
- [Computation-efficient Deep Learning for Computer Vision: A Survey](https://mlsanity.com/paper/2308.13998/) (2023)
- [One-Shot Neural Ensemble Architecture Search by Diversity-Guided Search Space Shrinking](https://mlsanity.com/paper/2104.00597/) (2021)
- [GreedyNASv2: Greedier Search with a Greedy Path Filter](https://mlsanity.com/paper/2111.12609/) (2021)

## Abstract (arXiv)

Training a supernet matters for one-shot neural architecture search (NAS) methods since it serves as a basic performance estimator for different architectures (paths). Current methods mainly hold the assumption that a supernet should give a reasonable ranking over all paths. They thus treat all paths equally, and spare much effort to train paths. However, it is harsh for a single supernet to evaluate accurately on such a huge-scale search space (e.g., 721). In this paper, instead of covering all paths, we ease the burden of supernet by encouraging it to focus more on evaluation of those potentially-good ones, which are identified using a surrogate portion of validation data. Concretely, during training, we propose a multi-path sampling strategy with rejection, and greedily filter the weak paths. The training efficiency is thus boosted since the training space has been greedily shrunk from all paths to those potentially-good ones. Moreover, we further adopt an exploration and exploitation policy by introducing an empirical candidate path pool. Our proposed method GreedyNAS is easy-to-follow, and experimental results on ImageNet dataset indicate that it can achieve better Top-1 accuracy under same search space and FLOPs or latency level, but with only ~60% of supernet training cost. By searching on a larger space, our GreedyNAS can also obtain new state-of-the-art architectures.