Predicting Neural Network Accuracy from Weights
| Autor: | Unterthiner, Thomas, Keysers, Daniel, Gelly, Sylvain, Bousquet, Olivier, Tolstikhin, Ilya |
|---|---|
| Rok vydání: | 2020 |
| Předmět: | |
| Druh dokumentu: | Working Paper |
| Popis: | 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. Comment: Updated the Small CNN Zoo dataset: reduced the maximal learning rate and got rid of multiple bad runs. Replaced all the experiments with the new numbers. Added MLP. Fixed typo in the abstract (R2 score instead of Kendall's tau). Added several earlier related works to the literature overview |
| Databáze: | arXiv |
| Externí odkaz: |
|