The accuracy and reproducibility of scientific discoveries made with machine-learning techniques should be questioned by scientists until systems can be developed that effectively critique themselves.
Sourced through Scoop.it from: www.biometricupdate.com
