Interpretable Machine Learning

Interpretable machine learning is potentially landmark research. Developed originally at Duke University and then at the University of Maine, ProtoPNet provides a means of resolving the mystery of how deep neural networks classify inputs.