DetOFA: Efficient Training of Once-for-All Networks for Object Detection Using Path FilterView Publication
We address the challenge of training a large supernet for the object detection task, using a relatively small amount of training data. Specifically, we propose an efficient supernetbased neural architecture search (NAS) method that uses search space pruning. The search space defined by the supernet is pruned by removing candidate models that are predicted to perform poorly. To effectively remove the candidates over a wide range of resource constraints, we particularly design a performance predictor, called path filter, which can accurately predict the relative performance of the models that satisfy similar resource constraints. Hence, supernet training is more focused on the best-performing candidates. Our path filter handles prediction for paths with different resource budgets. Compared to once-for-all, our proposed method reduces the computational cost of the optimal network architecture by 30% and 63%, while yielding better accuracy-floating point operations Pareto front (0.85 and 0.45 points of improvement on average precision for Pascal VOC and COCO, respectively). The code is attached to the supplementary materials.
Related PublicationsView All
Fine-grained Image Editing by Pixel-wise Guidance Using Diffusion Models
Naoki Matsunaga, Masato Ishii, Akio Hayakawa, Kenji Suzuki, Takuya NarihiraOur goal is to develop fine-grained real-image editing methods suitable for real-world applications. In this p […]
Data Cleansing for Deep Neural Networks with Storage-efficient Approximation of Influence Functions
Kenji Suzuki, Yoshiyuki Kobayashi, Takuya NarihiraIdentifying the influence of training data for data cleansing can improve the accuracy of deep learning. An ap […]