AtomNAS: Fine-Grained End-to-End Neural Architecture Search

TL;DR

AtomNAS introduces a fine-grained, atomic block-based search space for neural architecture search, enabling mixed operations and resource-aware optimization, resulting in state-of-the-art performance with efficient search on ImageNet.

Contribution

The paper presents a novel atomic block-based search space and a resource-aware, dynamic pruning framework for end-to-end neural architecture search.

Findings

Achieves state-of-the-art accuracy on ImageNet.

Reduces search cost compared to previous NAS methods.

Supports mixed operation architectures within the search space.

Abstract

Search space design is very critical to neural architecture search (NAS) algorithms. We propose a fine-grained search space comprised of atomic blocks, a minimal search unit that is much smaller than the ones used in recent NAS algorithms. This search space allows a mix of operations by composing different types of atomic blocks, while the search space in previous methods only allows homogeneous operations. Based on this search space, we propose a resource-aware architecture search framework which automatically assigns the computational resources (e.g., output channel numbers) for each operation by jointly considering the performance and the computational cost. In addition, to accelerate the search process, we propose a dynamic network shrinkage technique which prunes the atomic blocks with negligible influence on outputs on the fly. Instead of a search-and-retrain two-stage paradigm, our method simultaneously searches and trains the target architecture. Our method achieves state-of-the-art performance under several FLOPs configurations on ImageNet with a small searching cost. We open our entire codebase at: https://github.com/meijieru/AtomNAS.