v> Abstract Recently, Neural Architecture Search (NAS) has successfully identifified neural network architectures that exceed human designed ones on large-scale image classifification. In this paper, we study NAS for semantic image segmentation. Existing works often focus on searching the repeatable cell structure, while hand-designing the outer network structure that controls the spatial resolution changes. This choice simplififies the search space, but becomes increasingly problematic for dense image prediction which exhibits a lot more network level architectural variations. Therefore, we propose to search the network level structure in addition to the cell level structure, which forms a hierarchical architecture search space. We present a network level search space that includes many popular designs, and develop a formulation that allows effificient gradient-based architecture search (3 P100 GPU days on Cityscapes images). We demonstrate the effectiveness of the proposed method on the challenging Cityscapes, PASCAL VOC 2012, and ADE20K datasets. Auto-DeepLab, our architecture searched specififically for semantic image segmentation, attains state-of-the-art performance without any ImageNet pretraining.