Prioritized Architecture Sampling with Monto-Carlo Tree Search
arXiv:2103.11922
Abstract
One-shot neural architecture search (NAS) methods significantly reduce the search cost by considering the whole search space as one network, which only needs to be trained once. However, current methods select each operation independently without considering previous layers. Besides, the historical information obtained with huge computation cost is usually used only once and then discarded. In this paper, we introduce a sampling strategy based on Monte Carlo tree search (MCTS) with the search space modeled as a Monte Carlo tree (MCT), which captures the dependency among layers. Furthermore, intermediate results are stored in the MCT for the future decision and a better exploration-exploitation balance. Concretely, MCT is updated using the training loss as a reward to the architecture performance; for accurately evaluating the numerous nodes, we propose node communication and hierarchical node selection methods in the training and search stages, respectively, which make better uses of the operation rewards and hierarchical information. Moreover, for a fair comparison of different NAS methods, we construct an open-source NAS benchmark of a macro search space evaluated on CIFAR-10, namely NAS-Bench-Macro. Extensive experiments on NAS-Bench-Macro and ImageNet demonstrate that our method significantly improves search efficiency and performance. For example, by only searching architectures, our obtained architecture achieves top-1 accuracy with 442M FLOPs on ImageNet. Code (Benchmark) is available at: \url{https://github.com/xiusu/NAS-Bench-Macro}.
Accepted by CVPR2021. We also release a NAS benchmark on the one-shot MobileNetV2 search space
References in corpus (15)
- Distilling the Knowledge in a Neural Network
- EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
- MobileNetV2: Inverted Residuals and Linear Bottlenecks
- Transformer in Transformer
- ISTA-NAS: Efficient and Consistent Neural Architecture Search by Sparse Coding
- Sample-Efficient Neural Architecture Search by Learning Action Space
- Learning Search Space Partition for Black-box Optimization using Monte Carlo Tree Search
- PPDM: Parallel Point Detection and Matching for Real-time Human-Object Interaction Detection
- Locally Free Weight Sharing for Network Width Search
- Hit-Detector: Hierarchical Trinity Architecture Search for Object Detection
- Point-Set Anchors for Object Detection, Instance Segmentation and Pose Estimation
- BETANAS: BalancEd TrAining and selective drop for Neural Architecture Search
- GreedyNAS: Towards Fast One-Shot NAS with Greedy Supernet
- Towards Improving the Consistency, Efficiency, and Flexibility of Differentiable Neural Architecture Search
- Powering One-shot Topological NAS with Stabilized Share-parameter Proxy