AutoRF: Auto Learning Receptive Fields with Spatial Pooling

Peijie Dong, Xin Niu, Zimian Wei, Hengyue Pan, Dongsheng Li, Zhen Huang

Abstract: The search space is crucial in neural architecture search (NAS), and can determine the upper limit of the performance. Most methods focus on the design of depth and width when designing the search space, ignoring the receptive field. With a larger receptive field, the model is able to aggregate hierarchical information and strengthen its representational power. However, expanding the receptive fields directly with large convolution kernels suffers from high computational complexity. We instead enlarge the receptive field by introducing pooling operations with little overhead. In this paper, we propose a method named Auto Learning Receptive Fields (AutoRF), which is the first attempt at the auto attention module design with regard to the adaptive receptive field. In this paper, we present a pooling-based auto-learning approach for receptive field search. Our proposed search space encompasses typical multi-scale receptive field integration modules theoretically. Detailed experiments demonstrate the generalization ability of AutoRF and outperform various hand-crafted methods as well as NAS-based ones.