Predictions obtained by, eg, artificial neural networks have a high accuracy but humans
often perceive the models as black boxes. Insights about the decision making are mostly …

Recent developments in machine vision methods for automatic, quantitative analysis of
social behavior have immensely improved both the scale and level of resolution with which …

Dataset distillation compresses large datasets into smaller synthetic coresets which retain
performance with the aim of reducing the storage and computational burden of processing …

How can we explain the predictions of a black-box model? In this paper, we use influence
functions—a classic technique from robust statistics—to trace a model's prediction through …

We introduce TIDE, a framework and associated toolbox (https://dbolya. github. io/tide/) for
analyzing the sources of error in object detection and instance segmentation algorithms …

Post-hoc interpretability approaches have been proven to be powerful tools to generate
explanations for the predictions made by a trained black-box model. However, they create …

The great success of deep learning heavily relies on increasingly larger training data, which
comes at a price of huge computational and infrastructural costs. This poses crucial …

Data-trained predictive models see widespread use, but for the most part they are used as
black boxes which output a prediction or score. It is therefore hard to acquire a deeper …

In a world where traditional notions of privacy are increasingly challenged by the myriad
companies that collect and analyze our data, it is important that decision-making entities are …

In the context of post-hoc interpretability, this paper addresses the task of explaining the
prediction of a classifier, considering the case where no information is available, neither on …