Network Space Search for Pareto-Efficient Spaces

TL;DR

This paper introduces Network Space Search (NSS), a method to automatically discover efficient network spaces that are Pareto-efficient, reducing manual effort and improving neural architecture search outcomes.

Contribution

The paper proposes NSS to automatically search for favorable network spaces, creating Elite Spaces that are Pareto-efficient and suitable for NAS, with minimal human expertise.

Findings

Elite Spaces are Pareto-efficient and align with the Pareto front.

NSS reduces manual effort and search cost in NAS.

Achieves lower error rates and closer to constraints with fewer samples.

Abstract

Network spaces have been known as a critical factor in both handcrafted network designs or defining search spaces for Neural Architecture Search (NAS). However, an effective space involves tremendous prior knowledge and/or manual effort, and additional constraints are required to discover efficiency-aware architectures. In this paper, we define a new problem, Network Space Search (NSS), as searching for favorable network spaces instead of a single architecture. We propose an NSS method to directly search for efficient-aware network spaces automatically, reducing the manual effort and immense cost in discovering satisfactory ones. The resultant network spaces, named Elite Spaces, are discovered from Expanded Search Space with minimal human expertise imposed. The Pareto-efficient Elite Spaces are aligned with the Pareto front under various complexity constraints and can be further served as NAS search spaces, benefiting differentiable NAS approaches (e.g. In CIFAR-100, an averagely 2.3% lower error rate and 3.7% closer to target constraint than the baseline with around 90% fewer samples required to find satisfactory networks). Moreover, our NSS approach is capable of searching for superior spaces in future unexplored spaces, revealing great potential in searching for network spaces automatically. Website: https://minhungchen.netlify.app/publication/nss.