Making deep neural networks right for the right scientific reasons by interacting with their explanations

arXiv:2001.05371v4 Announce Type: replace
Abstract: Deep neural networks have shown excellent performances in many real-world applications. Unfortunately, they may show "Clever Hans"-like behavior -- making use of confounding factors within datasets -- to achieve high performance. In this work, we introduce the novel learning setting of "explanatory interactive learning" (XIL) and illustrate its benefits on a plant phenotyping research task. XIL adds the scientist into the training loop such that she interactively revises the original model via providing feedback on …

abstract applications arxiv behavior confounding cs.ai cs.lg datasets hans making networks neural networks novel performance performances show stat.ml type work world