In-Distribution Interpretability for Challenging Modalities
This work provides interpretability for AI models in specific domains like music and urban simulations, but it is incremental as it extends an existing method to new data.
The paper tackled the problem of interpreting deep neural network predictions by applying an existing generative model framework to music and urban simulation data, demonstrating its flexibility across challenging modalities.
It is widely recognized that the predictions of deep neural networks are difficult to parse relative to simpler approaches. However, the development of methods to investigate the mode of operation of such models has advanced rapidly in the past few years. Recent work introduced an intuitive framework which utilizes generative models to improve on the meaningfulness of such explanations. In this work, we display the flexibility of this method to interpret diverse and challenging modalities: music and physical simulations of urban environments.