Unmasking Clever Hans predictors and assessing what machines really learn
oleh: Sebastian Lapuschkin, Stephan Wäldchen, Alexander Binder, Grégoire Montavon, Wojciech Samek, Klaus-Robert Müller
| Format: | Article |
|---|---|
| Diterbitkan: | Nature Portfolio 2019-03-01 |
Deskripsi
Nonlinear machine learning methods have good predictive ability but the lack of transparency of the algorithms can limit their use. Here the authors investigate how these methods approach learning in order to assess the dependability of their decision making.