Stronger NAS with Weaker Predictors
arXiv:2102.10490
Abstract
Neural Architecture Search (NAS) often trains and evaluates a large number of architectures. Recent predictor-based NAS approaches attempt to alleviate such heavy computation costs with two key steps: sampling some architecture-performance pairs and fitting a proxy accuracy predictor. Given limited samples, these predictors, however, are far from accurate to locate top architectures due to the difficulty of fitting the huge search space. This paper reflects on a simple yet crucial question: if our final goal is to find the best architecture, do we really need to model the whole space well?. We propose a paradigm shift from fitting the whole architecture space using one strong predictor, to progressively fitting a search path towards the high-performance sub-space through a set of weaker predictors. As a key property of the weak predictors, their probabilities of sampling better architectures keep increasing. Hence we only sample a few well-performed architectures guided by the previously learned predictor and estimate a new better weak predictor. This embarrassingly easy framework, dubbed WeakNAS, produces coarse-to-fine iteration to gradually refine the ranking of sampling space. Extensive experiments demonstrate that WeakNAS costs fewer samples to find top-performance architectures on NAS-Bench-101 and NAS-Bench-201. Compared to state-of-the-art (SOTA) predictor-based NAS methods, WeakNAS outperforms all with notable margins, e.g., requiring at least 7.5x less samples to find global optimal on NAS-Bench-101. WeakNAS can also absorb their ideas to boost performance more. Further, WeakNAS strikes the new SOTA result of 81.3% in the ImageNet MobileNet Search Space. The code is available at https://github.com/VITA-Group/WeakNAS.
arXiv commentAccepted at NeurIPS 2021
References in corpus (22)
- Neural Architecture Search with Reinforcement Learning
- Neural Architecture Optimization
- PC-DARTS: Partial Channel Connections for Memory-Efficient Architecture Search
- A Downsampled Variant of ImageNet as an Alternative to the CIFAR datasets
- SNAS: Stochastic Neural Architecture Search
- NAS-Bench-101: Towards Reproducible Neural Architecture Search
- Single Path One-Shot Neural Architecture Search with Uniform Sampling
- BRP-NAS: Prediction-based NAS using GCNs
- Bridging the Gap between Sample-based and One-shot Neural Architecture Search with BONAS
- AlphaX: eXploring Neural Architectures with Deep Neural Networks and Monte Carlo Tree Search
- Semi-Supervised Neural Architecture Search
- A Generic Graph-based Neural Architecture Encoding Scheme for Predictor-based NAS
- DropNAS: Grouped Operation Dropout for Differentiable Architecture Search
- Convergence rates for an inexact ADMM applied to separable convex optimization
- FBNetV3: Joint Architecture-Recipe Search using Predictor Pretraining
- Neural Predictor for Neural Architecture Search
- ReNAS:Relativistic Evaluation of Neural Architecture Search
- Interpretable Neural Architecture Search via Bayesian Optimisation with Weisfeiler-Lehman Kernels
- Angle-based Search Space Shrinking for Neural Architecture Search
- BANANAS: Bayesian Optimization with Neural Architectures for Neural Architecture Search
- HourNAS: Extremely Fast Neural Architecture Search Through an Hourglass Lens
- DC-NAS: Divide-and-Conquer Neural Architecture Search