However,existingapproachesusually require alargesetofhuman annotated explanations fortraining while collecting a large set of explanations is not only time consuming but also expensive.

We initiate the study of behavioral information design through deep learning.

Now, we can exactly derive that q (G) = ˆ p( G|I) . The definitions of potential function φ and ψ follow those in JM-SGG model. Figure 1: The scene graphs generated by JM-SGG model. In these examples, factor update is able to correct some wrong relation labels ( e.g.

We carefully design the potential functions to enable relational reasoning among different objects according to knowledge graph embedding methods.

Optimizationwith Non-Differentiable Constraintswith Applicationsto Fairness, Recall, Churn, and Other Goals.Journalof Machine Learning Research, 20(172): 1-59, 2019.

We consider the problem of hypothesis testing for discrete distributions.

(Citedon 16)