BANANAS: Bayesian Optimization with Neural Architectures for Neural Architecture Search

Abstract

Over the past half-decade, many methods have been considered for neural architecture search (NAS). Bayesian optimization (BO), which has long had success in hyperparameter optimization, has recently emerged as a very promising strategy for NAS when it is coupled with a neural predictor. Recent work has proposed different instantiations of this framework, for example, using Bayesian neural networks or graph convolutional networks as the predictive model within BO. However, the analyses in these papers often focus on the full-fledged NAS algorithm, so it is difficult to tell which individual components of the framework lead to the best performance. In this work, we give a thorough analysis of the "BO + neural predictor" framework by identifying five main components: the architecture encoding, neural predictor, uncertainty calibration method, acquisition function, and acquisition optimization strategy. We test several different methods for each component and also develop a novel path-based encoding scheme for neural architectures, which we show theoretically and empirically scales better than other encodings. Using all of our analyses, we develop a final algorithm called BANANAS, which achieves state-of-the-art performance on NAS search spaces. We adhere to the NAS research checklist (Lindauer and Hutter 2019) to facilitate best practices, and our code is available at https://github.com/naszilla/naszilla.

Cited by in corpus (9)

• How Powerful are Performance Predictors in Neural Architecture Search?
• Stronger NAS with Weaker Predictors
• Optimal Transport Kernels for Sequential and Parallel Neural Architecture Search
• CATE: Computation-aware Neural Architecture Encoding with Transformers
• DAAS: Differentiable Architecture and Augmentation Policy Search
• NAS-HPO-Bench-II: A Benchmark Dataset on Joint Optimization of Convolutional Neural Network Architecture and Training Hyperparameters
• ProxyBO: Accelerating Neural Architecture Search via Bayesian Optimization with Zero-cost Proxies
• Exploring the Loss Landscape in Neural Architecture Search
• Training BatchNorm Only in Neural Architecture Search and Beyond