论文标题
预测重量的神经网络准确性
Predicting Neural Network Accuracy from Weights
论文作者
论文摘要
我们通过实验表明，可以通过仅查看其权重，而无需对输入数据进行评估，可以很好地预测受过训练的神经网络的准确性。我们激励这项任务，并为其引入正式设置。即使使用权重的简单统计数据，预测因子也能够以非常高的精度来对神经网络进行排名（R2得分超过0.98）。此外，这些预测因素能够对在不同，未观察到的数据集和不同体系结构进行训练的网络进行排名。我们发布了在四个不同数据集中训练的120k卷积神经网络的集合，以鼓励在该领域进行进一步的研究，以便更好地了解网络培训和性能。
We show experimentally that the accuracy of a trained neural network can be predicted surprisingly well by looking only at its weights, without evaluating it on input data. We motivate this task and introduce a formal setting for it. Even when using simple statistics of the weights, the predictors are able to rank neural networks by their performance with very high accuracy (R2 score more than 0.98). Furthermore, the predictors are able to rank networks trained on different, unobserved datasets and with different architectures. We release a collection of 120k convolutional neural networks trained on four different datasets to encourage further research in this area, with the goal of understanding network training and performance better.
