Landmark Regularization: Ranking Guided Super-Net Training in Neural Architecture Search
This addresses a critical bottleneck in neural architecture search for researchers and practitioners, though it is incremental as it builds on existing weight-sharing methods.
The paper tackled the ranking disorder problem in weight-sharing neural architecture search by introducing a regularization term that maximizes correlation between shared-weight and standalone architecture performance rankings, showing consistent performance improvements across algorithms, search-spaces, and tasks.
Weight sharing has become a de facto standard in neural architecture search because it enables the search to be done on commodity hardware. However, recent works have empirically shown a ranking disorder between the performance of stand-alone architectures and that of the corresponding shared-weight networks. This violates the main assumption of weight-sharing NAS algorithms, thus limiting their effectiveness. We tackle this issue by proposing a regularization term that aims to maximize the correlation between the performance rankings of the shared-weight network and that of the standalone architectures using a small set of landmark architectures. We incorporate our regularization term into three different NAS algorithms and show that it consistently improves performance across algorithms, search-spaces, and tasks.