The design of neural network architectures for a new data set is a laborious
task which requires human deep learning expertise. In order to make deep
learning available for a broader audience, automated methods for finding a
neural network architecture are vital. Recently proposed meth
本文提出了一个名为 AlphaX 的新型可扩展 Monte Carlo Tree Search (MCTS) NAS 代理,该代理通过元深度神经网络 (DNN) 预测网络准确度,进而优化探索效率,降低网络评估成本,且在 CIFAR-10、ImageNet 和 NASBench-101 数据集上均取得了优异的结果。