Interpretations are useful: penalizing explanations to align neural networks with prior knowledge

Add code
Oct 01, 2019
Figure 1 for Interpretations are useful: penalizing explanations to align neural networks with prior knowledge
Figure 2 for Interpretations are useful: penalizing explanations to align neural networks with prior knowledge
Figure 3 for Interpretations are useful: penalizing explanations to align neural networks with prior knowledge
Figure 4 for Interpretations are useful: penalizing explanations to align neural networks with prior knowledge

Share this with someone who'll enjoy it:

View paper onarxiv icon

Share this with someone who'll enjoy it: