Incorporating Domain Knowledge into Deep Neural Networks
It addresses the problem of enhancing model construction through human-machine collaboration, but it is incremental as it primarily reviews existing approaches.
This paper surveys methods for incorporating domain knowledge into deep neural networks, focusing on logical and numerical constraints, and reviews techniques and results across various sub-categories.
We present a survey of ways in which domain-knowledge has been included when constructing models with neural networks. The inclusion of domain-knowledge is of special interest not just to constructing scientific assistants, but also, many other areas that involve understanding data using human-machine collaboration. In many such instances, machine-based model construction may benefit significantly from being provided with human-knowledge of the domain encoded in a sufficiently precise form. This paper examines two broad approaches to encode such knowledge--as logical and numerical constraints--and describes techniques and results obtained in several sub-categories under each of these approaches.