Guarded Deep Learning using Scenario-Based Modeling
This addresses safety concerns for users deploying opaque DNN models, but is incremental as it builds on existing scenario-based modeling.
The paper tackles the problem of making deep neural networks safer and more interpretable by augmenting them with override rules expressed as intuitive scenarios, resulting in increased overall safety without specifying concrete numbers.
Deep neural networks (DNNs) are becoming prevalent, often outperforming manually-created systems. Unfortunately, DNN models are opaque to humans, and may behave in unexpected ways when deployed. One approach for allowing safer deployment of DNN models calls for augmenting them with hand-crafted override rules, which serve to override decisions made by the DNN model when certain criteria are met. Here, we propose to bring together DNNs and the well-studied scenario-based modeling paradigm, by expressing these override rules as simple and intuitive scenarios. This approach can lead to override rules that are comprehensible to humans, but are also sufficiently expressive and powerful to increase the overall safety of the model. We describe how to extend and apply scenario-based modeling to this new setting, and demonstrate our proposed technique on multiple DNN models.