Guided Evolution for Neural Architecture Search
This addresses the challenge of efficient and effective NAS for researchers and practitioners in computer vision, though it appears incremental as it builds on existing evolutionary methods with a novel guidance mechanism.
The paper tackles the problem of Neural Architecture Search (NAS) methods being complex and prone to local minima by proposing G-EA, a guided evolutionary approach that uses a zero-proxy estimator to explore the search space efficiently, achieving state-of-the-art results with mean accuracies of 93.98% on CIFAR-10, 72.12% on CIFAR-100, and 45.94% on ImageNet16-120.
Neural Architecture Search (NAS) methods have been successfully applied to image tasks with excellent results. However, NAS methods are often complex and tend to converge to local minima as soon as generated architectures seem to yield good results. In this paper, we propose G-EA, a novel approach for guided evolutionary NAS. The rationale behind G-EA, is to explore the search space by generating and evaluating several architectures in each generation at initialization stage using a zero-proxy estimator, where only the highest-scoring network is trained and kept for the next generation. This evaluation at initialization stage allows continuous extraction of knowledge from the search space without increasing computation, thus allowing the search to be efficiently guided. Moreover, G-EA forces exploitation of the most performant networks by descendant generation while at the same time forcing exploration by parent mutation and by favouring younger architectures to the detriment of older ones. Experimental results demonstrate the effectiveness of the proposed method, showing that G-EA achieves state-of-the-art results in NAS-Bench-201 search space in CIFAR-10, CIFAR-100 and ImageNet16-120, with mean accuracies of 93.98%, 72.12% and 45.94% respectively.