Abstract | ||
---|---|---|
Evaluating a neural network on an input that differs markedly from the training data might cause erratic and flawed predictions. We study a method that judges the unusualness of an input by evaluating its informative content compared to the learned parameters. This technique can be used to judge whether a network is suitable for processing a certain input and to raise a red flag that unexpected behavior might lie ahead. We compare our approach to various methods for uncertainty evaluation from the literature for various datasets and scenarios. Specifically, we introduce a simple, effective method that allows to directly compare the output of such metrics for single input points even if these metrics live on different scales. |
Year | DOI | Venue |
---|---|---|
2021 | 10.1007/s10489-020-01925-8 | APPLIED INTELLIGENCE |
Keywords | DocType | Volume |
Deep learning, Trustworthiness, Fisher information, Uncertainty, Out-of-distribution | Journal | 51 |
Issue | ISSN | Citations |
4 | 0924-669X | 0 |
PageRank | References | Authors |
0.34 | 0 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Jörg Martin | 1 | 0 | 0.68 |
Clemens Elster | 2 | 96 | 14.27 |