GeNAS: Neural Architecture Search with Better Generalization
This work addresses the challenge of improving generalization in neural architecture search for machine learning practitioners, though it is incremental as it builds on existing NAS approaches with a new measure.
The paper tackles the problem of finding neural architectures with better generalization by proposing a new measure based on the flatness of the loss surface, showing that it achieves similar or better performance than state-of-the-art NAS methods and robustly generalizes to data distribution shifts and various tasks.
Neural Architecture Search (NAS) aims to automatically excavate the optimal network architecture with superior test performance. Recent neural architecture search (NAS) approaches rely on validation loss or accuracy to find the superior network for the target data. In this paper, we investigate a new neural architecture search measure for excavating architectures with better generalization. We demonstrate that the flatness of the loss surface can be a promising proxy for predicting the generalization capability of neural network architectures. We evaluate our proposed method on various search spaces, showing similar or even better performance compared to the state-of-the-art NAS methods. Notably, the resultant architecture found by flatness measure generalizes robustly to various shifts in data distribution (e.g. ImageNet-V2,-A,-O), as well as various tasks such as object detection and semantic segmentation. Code is available at https://github.com/clovaai/GeNAS.