On Network Design Spaces for Visual Recognition
This work addresses the methodology for comparing network architectures, which is crucial for researchers in computer vision and machine learning to sustain progress in designing better models, though it is incremental in refining comparison techniques.
The authors tackled the problem of comparing neural network architectures for visual recognition by introducing a new comparison paradigm based on distribution estimates, which revealed significant statistical differences between design spaces and showed that standard model families can be comparable to more complex ones used in neural architecture search.
Over the past several years progress in designing better neural network architectures for visual recognition has been substantial. To help sustain this rate of progress, in this work we propose to reexamine the methodology for comparing network architectures. In particular, we introduce a new comparison paradigm of distribution estimates, in which network design spaces are compared by applying statistical techniques to populations of sampled models, while controlling for confounding factors like network complexity. Compared to current methodologies of comparing point and curve estimates of model families, distribution estimates paint a more complete picture of the entire design landscape. As a case study, we examine design spaces used in neural architecture search (NAS). We find significant statistical differences between recent NAS design space variants that have been largely overlooked. Furthermore, our analysis reveals that the design spaces for standard model families like ResNeXt can be comparable to the more complex ones used in recent NAS work. We hope these insights into distribution analysis will enable more robust progress toward discovering better networks for visual recognition.