Unmasking Clever Hans predictors and assessing what machines really learn

Current learning machines have successfully solved hard application problems, reaching high accuracy and displaying seemingly intelligent behavior. Here we apply recent techniques for explaining decisions of state-of-the-art learning machines and analyze various tasks from computer vision and arcade games. This showcases a spectrum of problem-solving behaviors ranging from naive and short-sighted, to well-informed and strategic. We observe that standard performance evaluation metrics can be oblivious to distinguishing these diverse problem solving behaviors. Furthermore, we propose our semi-automated Spectral Relevance Analysis that provides a practically effective way of characterizing and validating the…

• DF
  Deutsche Forschungsgemeinschaft

  Awards: EXC 2046, 01IS18037I, EXC 2046/1, project-ID: 390685689, project-ID: 390685689, 390685689, EXC 2046/1
• IF
  Institute for Information and Communications Technology Promotion

  Award: 2017-0-00451